I've been pacing myself to two pages added on to my thesis per day. This was a very good suggestion by one of my advisers, it is enough to feel as if I did something that day and not so much to feel burn out by the work to get those two pages. Also, if I continue at this pace, I will definitely be able to reach my goals of a completed thesis (minus revisions) by the end of May. Early June I will spend a week making a powerpoint presentation and perhaps defend in late June. This would give me time to make any revisions my committe suggest and turn in my thesis by the deadline July 13.
Well I am basically finished with my two pages so I will go home. Tomorrow I will meet with my adviser in the morning to outline of the remaining chapters of my thesis; I also need to assign final grades for my students.
Thursday, April 30, 2009
Numb3rs
Recently I started watching the show numb3rs again. Briefly it is a story about a genius mathematician that uses his quantitative skills to solve crimes being investigated by the FBI.
When this show first came out, I was very excited to have a math show so popular. Finally everyday people would get a chance to peer into the many applications of complex mathematics. Even to this day, I appreciate the effort at least. However, I must say after watching a few episodes of season three, I almost had to cringe at the ridiculousness of it all.
It has become quite clear that whoever the producer is does not really care about making the math make any kind of sense at all. Sure they consults with mathematicians, that do give him real math equations, but the applications are very unrealistic at times.
At typical storyline will go something like this:
FBI: We have two suspects somewhere in Miami that have a bomb, how can we find them?
Math Guy: We can use Quantum Theory along with fractal geometry to create a high probability density as to where they might be. Give me 1 minute....here is a list of five locations with the highest probability of their location.
FBI: Great, we found them at location 4...your awesome!
What's wrong with this picture? I guess the show is hoping that once the Math Guy starts speaking in "math" that they will assume he is doing some math tricks they don't understand but can actually give you this sort of information. But common, math is not magic...sure you can apply it to these types of problem but this problem is too general to yield any meaningful results.
As a math person that watches the show, I don't expect to understand the math or even all it's applications. But I would like to believe that what they are doing on the show is actually plausible; I would like to also believe that this math 'could' be done. I have come to the conclusion that I must not be their target audience.
When this show first came out, I was very excited to have a math show so popular. Finally everyday people would get a chance to peer into the many applications of complex mathematics. Even to this day, I appreciate the effort at least. However, I must say after watching a few episodes of season three, I almost had to cringe at the ridiculousness of it all.
It has become quite clear that whoever the producer is does not really care about making the math make any kind of sense at all. Sure they consults with mathematicians, that do give him real math equations, but the applications are very unrealistic at times.
At typical storyline will go something like this:
FBI: We have two suspects somewhere in Miami that have a bomb, how can we find them?
Math Guy: We can use Quantum Theory along with fractal geometry to create a high probability density as to where they might be. Give me 1 minute....here is a list of five locations with the highest probability of their location.
FBI: Great, we found them at location 4...your awesome!
What's wrong with this picture? I guess the show is hoping that once the Math Guy starts speaking in "math" that they will assume he is doing some math tricks they don't understand but can actually give you this sort of information. But common, math is not magic...sure you can apply it to these types of problem but this problem is too general to yield any meaningful results.
As a math person that watches the show, I don't expect to understand the math or even all it's applications. But I would like to believe that what they are doing on the show is actually plausible; I would like to also believe that this math 'could' be done. I have come to the conclusion that I must not be their target audience.
Wednesday, April 29, 2009
Bad news
My computer adapters is starting to have the same problems my old computer had. I knew I should have taken out the warranty! I have never really liked this computer but I bought it to just get some work done without really looking into the specs. Shortly after purchasing it a knob fell off one of the keys and I almost lost another one as well.
I will probably go out an buy another laptop, this time paying more attention to the specs and obtaining at least a two year warranty. I went back to best buy to look at laptops but I think I might get one from walmart. Their 2 year warranty is under $100 while Best Buy's is $300.
I took some time off from look yesterday afternoon as I've been feeling pressure behind my left eye and I figured I needed a break. I went to borders and tried to buy a book with my credit card only to have it get declined. Apparently, credit card companies have been deactivation cards that are not being used and have zero balance. Luckily I had cash on me so I still bought the book.
I finished cleaning my room and trashing lots of things, and bought old clothes and shoes to goodwill. I also did a substantial amount of laundry and vacuumed. Once I was satisfied with the condition of my room I ate and starting reading my book. It is very rare for a book to engross me anymore. Any books I read I usually highly technical in nature. The last book I read for entertainment was about the Supreme Court Justices which is relatively less technical but not fiction reading.
The only fiction books I can even recall enjoying were the Dan Brown novels. I am anticipating the release of his latest book in the fall. Anyhow, I am reading a book about a missionaries experience living with the pihara tribe in the amazon jungle. This is the second time I have read a book about someone documenting their experiences with a tribal community. I believe the anthropologists in me finds it intriguing how some cultures have no concept of numbers, to them you either have one thing or many things. They don't even have a real concept of color. The interesting thing is that the missionary went there to convert the people to christianity and ended up realizing that their perception of reality was so different than his european understanding of life and death. I have only read the first few chapters but so far it's captured my interest.
Actually I did just finish reading another book about a black brain surgeon Keith Black. The book was interesting because he dealt alot with people given months to live. One thing that stuck with me is that he notices that people can actually live a full life in a few months; most of us never get around to this because we are so busy with life and expect to have so much more time to live in the future. It made me think about how my life would change if I were diagnosed with terminal brain cancer and only given a few months to live. It also made me want to live healthier!
I will probably go out an buy another laptop, this time paying more attention to the specs and obtaining at least a two year warranty. I went back to best buy to look at laptops but I think I might get one from walmart. Their 2 year warranty is under $100 while Best Buy's is $300.
I took some time off from look yesterday afternoon as I've been feeling pressure behind my left eye and I figured I needed a break. I went to borders and tried to buy a book with my credit card only to have it get declined. Apparently, credit card companies have been deactivation cards that are not being used and have zero balance. Luckily I had cash on me so I still bought the book.
I finished cleaning my room and trashing lots of things, and bought old clothes and shoes to goodwill. I also did a substantial amount of laundry and vacuumed. Once I was satisfied with the condition of my room I ate and starting reading my book. It is very rare for a book to engross me anymore. Any books I read I usually highly technical in nature. The last book I read for entertainment was about the Supreme Court Justices which is relatively less technical but not fiction reading.
The only fiction books I can even recall enjoying were the Dan Brown novels. I am anticipating the release of his latest book in the fall. Anyhow, I am reading a book about a missionaries experience living with the pihara tribe in the amazon jungle. This is the second time I have read a book about someone documenting their experiences with a tribal community. I believe the anthropologists in me finds it intriguing how some cultures have no concept of numbers, to them you either have one thing or many things. They don't even have a real concept of color. The interesting thing is that the missionary went there to convert the people to christianity and ended up realizing that their perception of reality was so different than his european understanding of life and death. I have only read the first few chapters but so far it's captured my interest.
Actually I did just finish reading another book about a black brain surgeon Keith Black. The book was interesting because he dealt alot with people given months to live. One thing that stuck with me is that he notices that people can actually live a full life in a few months; most of us never get around to this because we are so busy with life and expect to have so much more time to live in the future. It made me think about how my life would change if I were diagnosed with terminal brain cancer and only given a few months to live. It also made me want to live healthier!
Monday, April 27, 2009
First day of last week
Today is the first day of finals, and I will be giving mine from 3-5. I will be very glad to get this class over with. The moment I enter and submit finals grades, I will definitely breath a hug sigh of relief. I get very disorganized with my class toward the end, and it is reaching the tipping point. I have had several student contact me over a various number of issues and it is getting annoying.
Over the weekend I successfully applied my methodology to a electricity dataset. This was actually much further than I planned on going in my research, but since it fit so well and it didn't take long to do, I think it will be worth putting in my thesis. I also spent some time marveling at the robustness of the algorithm. It appears these results can identify a very general class of time series models with few assumptions. I am laying out an outline for the second writeup in my thesis, right now I believe it will go as follows:
1. Nonlinear Estimation Theory: particle filters, smoothers, improving the smoothing time
2. Nonlinear Parameter Estimation: Simple Model Filtering estimation, complex model with jumps, complex model filtering estimation.
3. Application: Simulated data, actual datasets
The good news is that I have a good skeleton writeup already and I will just build from it. My final thesis should be somewhere around 100 pages which will be more than sufficient.
I got accepted to the IMSM 2009 conference (my back up plan if I didnt get the FPL job) and got my July 20 start date confirmed by FPL this morning. I would like to do both, but the conference just so happens to be the week of my start date, and I wouldn't want to push that start date back over that. This reminds me that I need to speak with Pam since the last day to defend in Summer A might be in early July...This means that it's more likely my adviser will have me do the defense over skype but we will see.
Over the weekend I successfully applied my methodology to a electricity dataset. This was actually much further than I planned on going in my research, but since it fit so well and it didn't take long to do, I think it will be worth putting in my thesis. I also spent some time marveling at the robustness of the algorithm. It appears these results can identify a very general class of time series models with few assumptions. I am laying out an outline for the second writeup in my thesis, right now I believe it will go as follows:
1. Nonlinear Estimation Theory: particle filters, smoothers, improving the smoothing time
2. Nonlinear Parameter Estimation: Simple Model Filtering estimation, complex model with jumps, complex model filtering estimation.
3. Application: Simulated data, actual datasets
The good news is that I have a good skeleton writeup already and I will just build from it. My final thesis should be somewhere around 100 pages which will be more than sufficient.
I got accepted to the IMSM 2009 conference (my back up plan if I didnt get the FPL job) and got my July 20 start date confirmed by FPL this morning. I would like to do both, but the conference just so happens to be the week of my start date, and I wouldn't want to push that start date back over that. This reminds me that I need to speak with Pam since the last day to defend in Summer A might be in early July...This means that it's more likely my adviser will have me do the defense over skype but we will see.
Saturday, April 25, 2009
Very productive week
After my breakthrough earlier this weeks, my results and understanding of the problem have exponentially improved. My goal is to completely finish my write up on the old material this month so I can just focus on this new stuff which how now become quite interesting with many implications and promise.
I spoke with my adviser yesterday and one thing that I took from the meeting was the need to ensure that the work goes on. This is now added to my list of things to do, and I took one of my other colleagues to lunch yesterday so we could discuss her possibly pursuing some of this work. I personally think that it is a win-win for any graduate student with vision. I have topics...topics alone can take years to find. I know the papers you need to read and I know exactly what you need to know so that advisers will work with you. So if a new graduate student has the interest and motivation, they should have no problem whatsoever graduating.
My hope is to have some working group established in this area that can pursue the technical stuff (programming and theory) while I focus less on theory and programming and more on collecting data and suggesting models for the theory from my work. If we can create a productive working group I have no doubt we can all publish papers with minimal work.
Finishing my old work has been quite a pain, I've lost interest in most of it partly because I have talked about it so much and also because it doesn't seem as difficult as it once did. I wonder if my current work will feel the same one day.
I spoke with my adviser yesterday and one thing that I took from the meeting was the need to ensure that the work goes on. This is now added to my list of things to do, and I took one of my other colleagues to lunch yesterday so we could discuss her possibly pursuing some of this work. I personally think that it is a win-win for any graduate student with vision. I have topics...topics alone can take years to find. I know the papers you need to read and I know exactly what you need to know so that advisers will work with you. So if a new graduate student has the interest and motivation, they should have no problem whatsoever graduating.
My hope is to have some working group established in this area that can pursue the technical stuff (programming and theory) while I focus less on theory and programming and more on collecting data and suggesting models for the theory from my work. If we can create a productive working group I have no doubt we can all publish papers with minimal work.
Finishing my old work has been quite a pain, I've lost interest in most of it partly because I have talked about it so much and also because it doesn't seem as difficult as it once did. I wonder if my current work will feel the same one day.
Thursday, April 23, 2009
More Time..Less Stress
I was able to get my start date pushed back to July 20 and they said it would even be fine if I need longer. I said a week at the most. This works out perfect for me since my lease is up the beginning of August anyway and it basically give me another month to organize my thesis.
My work has been going quite well, ever since I stumbled upon that idea of reducing the jumps size. I have been able to run tests to diagnose the problems that were occurring earlier. I am varying the parameter slightly to see how the procedure reacts.
I have now spoken with all my committee members and they are all on board with my plans to defend this summer. The only issue now is finding a date that fits within everyone's schedule. Several professors have suggested being out of town for several weeks over the summer; so as a statistician I should probably calculate the probability that they will all be in town at the same time. Right now, I am thinking July 17 sounds good. My adviser will be back from France even though he said he could also do it via skype.
Apparently my thesis has been making it rounds around campus. My brother told me yesterday one of his students got a copy of it from the finance department and another graduate students in the statistics department is reading a copy to start his own research. I told him I will help catch him up to speed on my work over the summer and I am actually pleased that someone in our department will continue pursuing this line of work. There are many areas in the particle filter, smoothing literature that I just wont have time to explore but would no doubt yield a dissertation for someone willing to pursue it. There are also many applications in finance and economics that I see.
It's encouraging in the sense that currently I am the only graduate student in the department working in this area. Hopefully I have started some spark that will grow, and hope is that I can still do some collaboration with the other graduate student while I'm in S. Florida and perhaps get a paper out. It's not that I haven't tried to market my work before; however the graduate students I talked to either had areas of research or simply just didn't have the skills necessary to do the work. But this guy seems like he's willing to put in the work necessary to understand whats going on.
I won't go into details, but one lesson I learned yesterday from watching a colleagues essay defense. Never ever...ever put stuff on you slides that you are not prepared to defend.
I had my last class yesterday, after 3 years it feel like I'm leaving at the right time.
My work has been going quite well, ever since I stumbled upon that idea of reducing the jumps size. I have been able to run tests to diagnose the problems that were occurring earlier. I am varying the parameter slightly to see how the procedure reacts.
I have now spoken with all my committee members and they are all on board with my plans to defend this summer. The only issue now is finding a date that fits within everyone's schedule. Several professors have suggested being out of town for several weeks over the summer; so as a statistician I should probably calculate the probability that they will all be in town at the same time. Right now, I am thinking July 17 sounds good. My adviser will be back from France even though he said he could also do it via skype.
Apparently my thesis has been making it rounds around campus. My brother told me yesterday one of his students got a copy of it from the finance department and another graduate students in the statistics department is reading a copy to start his own research. I told him I will help catch him up to speed on my work over the summer and I am actually pleased that someone in our department will continue pursuing this line of work. There are many areas in the particle filter, smoothing literature that I just wont have time to explore but would no doubt yield a dissertation for someone willing to pursue it. There are also many applications in finance and economics that I see.
It's encouraging in the sense that currently I am the only graduate student in the department working in this area. Hopefully I have started some spark that will grow, and hope is that I can still do some collaboration with the other graduate student while I'm in S. Florida and perhaps get a paper out. It's not that I haven't tried to market my work before; however the graduate students I talked to either had areas of research or simply just didn't have the skills necessary to do the work. But this guy seems like he's willing to put in the work necessary to understand whats going on.
I won't go into details, but one lesson I learned yesterday from watching a colleagues essay defense. Never ever...ever put stuff on you slides that you are not prepared to defend.
I had my last class yesterday, after 3 years it feel like I'm leaving at the right time.
Wednesday, April 22, 2009
Stumbled upon
I stumbled upon a result today that verifies that my theory works! This idea came at just the right time. I had just gotten out of a meeting with my adviser in which we were discussing strategies on how to test the theory, some of the ideas were to run simulations that could take weeks, because all of my preliminary results suggested I needed a ridiculously large sample size.
The filtering technique definitely improved the time of my results but still the tests were going to consume alot of precious time without even knowing if the theory was correct. As I was going back to my office thinking about how much work I would have to do just to test the theory, I thought "what if I just decrease the jump size"?
I had done this before in some early tests but I didn't think to do it in the parameter estimation, however the logic suggest a smaller jump size would imply smaller particle size. Woila it works! The likelihood increases at each iteration which is consistent with the theory, my parameter estimates are not the same, but the theory doesnt guarantee that; only an increase in the likelihood. But this means at the very least that the theory works.
I cannot overestimate how much a sigh of relief this was, I have found dry ground. I barely know where I am, how I got here, or what this means, but I know I have dry ground. Right now I am running the algorithm from different initial conditions. This should be the beginning of the end.
The filtering technique definitely improved the time of my results but still the tests were going to consume alot of precious time without even knowing if the theory was correct. As I was going back to my office thinking about how much work I would have to do just to test the theory, I thought "what if I just decrease the jump size"?
I had done this before in some early tests but I didn't think to do it in the parameter estimation, however the logic suggest a smaller jump size would imply smaller particle size. Woila it works! The likelihood increases at each iteration which is consistent with the theory, my parameter estimates are not the same, but the theory doesnt guarantee that; only an increase in the likelihood. But this means at the very least that the theory works.
I cannot overestimate how much a sigh of relief this was, I have found dry ground. I barely know where I am, how I got here, or what this means, but I know I have dry ground. Right now I am running the algorithm from different initial conditions. This should be the beginning of the end.
Tuesday, April 21, 2009
Saving time
I believe I have discovered a way to save a bunch of computation time. This is actually something I already knew but I just hadnt applied it to this specific computation. Basically, the smoother takes about 8 hours to do 500 particle over 100 time steps...the filter can do the same computation in less than 2 minutes.
The cost is that the smoother has a better fit of the data, how much better of a fit? Not much...maybe a few percentage points better. This makes the filter desirable to at least test the accuracy of the computations. I am not sure if I can do parameter learning with it because it goes against the theory, but I will still try. I am currently testing the filter with 5000 particles...this would no doubt take about a month to do with the smoother. I will probably have very similar results in about an hour with the filter.
This is the accomplishment of the day. Once again, just scaling back the ambition for a while, making sure all my expectation computations are correct.
The cost is that the smoother has a better fit of the data, how much better of a fit? Not much...maybe a few percentage points better. This makes the filter desirable to at least test the accuracy of the computations. I am not sure if I can do parameter learning with it because it goes against the theory, but I will still try. I am currently testing the filter with 5000 particles...this would no doubt take about a month to do with the smoother. I will probably have very similar results in about an hour with the filter.
This is the accomplishment of the day. Once again, just scaling back the ambition for a while, making sure all my expectation computations are correct.
Monday, April 20, 2009
goal
My goal right now is to be 100% confidence that my expectation work. Forget the parameter estimation for now, I will scale down those ambition for another day. Currently programming tests for achieve this. So far the results are mixed.
Growing frustration
This problem is becoming more of a pain than and interest every day. The problem with this is that it makes me more prone to error when I am anxious and irritated with having to debug and reprogram stuff. I found another major error today in one subroutine that was inputting the wrong parameters.
This was probably something that I should have recognized earlier. Specifically since I noticed I was getting negative values in the output of the subroutine when they should clearly have been positive. After correcting this I made some progress in my computations. Still I have yet to achieve good results in the full routine.
My students have their presentations tomorrow and I really haven't given it much thought to how I will implement all of the presentations. Once again I will probably wait till the last minute. I had an hour long conversation with the relocation department for fpl. She basically laid out my relocation benefits and for some reason they still have me supposedly starting on may 11. Definitely not gonna happen.
I am having trouble locating one of my committee members. He seems to be out of his office every time I stop by. Hopefully I can catch him sometime this week.
This was probably something that I should have recognized earlier. Specifically since I noticed I was getting negative values in the output of the subroutine when they should clearly have been positive. After correcting this I made some progress in my computations. Still I have yet to achieve good results in the full routine.
My students have their presentations tomorrow and I really haven't given it much thought to how I will implement all of the presentations. Once again I will probably wait till the last minute. I had an hour long conversation with the relocation department for fpl. She basically laid out my relocation benefits and for some reason they still have me supposedly starting on may 11. Definitely not gonna happen.
I am having trouble locating one of my committee members. He seems to be out of his office every time I stop by. Hopefully I can catch him sometime this week.
Saturday, April 18, 2009
Week of events
I haven't written recently partially due to my extensive travels and hectic schedule. NC was good, I met a lot of nice people and the experience as a whole was enjoyable. The conference itself was sortof a disappointment. Not that anything was wrong with the conference, it was just that I was expecting more talk on sequential monte carlo method techniques but it was more on computer design of experiments.
Also, when I arrived I looked on the schedule and realized that I was listed as a poster presentation. So during the first couple of hours of the conference, I was on my laptop creating slides to put on a poster and planning to give a brief talk to a room full of statisticians! I winged it pretty successfully and got positive feedback about my work.
The research triangle in NC is an interesting place. There are no stores anywhere around and it definitely has the feel of a place in which you can spend time thinking deeply about many things. I found my short stay there very enjoyable, but I do believe I would become extremely bored there during an extended stay. At one point in my life, I probably would have enjoyed this seclusion more, but I have grown to enjoy the interaction of people and business of life outside of work.
I mistakenly booked my flight for the am instead of the pm which meant that I missed the last day of the conference which was friday. It was too expensive to change my ticket and I actually didn't mind once I realize the talks were not going to be very interesting to me. In fact, I spent most of the conference talk time just working on my laptop and running simulations for my thesis.
I discovered a bug in my many lines of code today. One that was very detrimental to my program and that I had overlooked several time due to it's subtle nature. The bug was so subtle that it was able to mask it self well when I ran my program in simple situation so that everything worked perfectly. It was only when I added "jumps" ty my model, did the but switch into effect.
Most of the problems that I run into actually have nothing to do with the theory I'm thinking being wrong, but rather the limiting power of the computer. For example, the computer might say I am dividing something by zero, which theoretically cant be zero but is so small the computer just says that it is.
Anyways this bug has moved me back into the direction of solving the problem that I had only recently concluded couldnt be solved. That conclusion would have been correct had I not found this bug. I will have to run more tests to be sure again.
This week my department celebrated it's 50 year anniversary. I got to meet up with several of my old colleague that have graduated already and the event was very nice. It's almost like a second family meeting when a group of statisticians are in a room together, there's a bond of jargon, and common purpose. The take raw data which can be very ugly and worthless and make something meaningful, powerful, and beautiful at times.
I missed the entire second day of events as I have been working intently to solve this problem in my office. I am setting a mental clock of two week for this problem; if it is not solved by then I will just give it up.
Also, when I arrived I looked on the schedule and realized that I was listed as a poster presentation. So during the first couple of hours of the conference, I was on my laptop creating slides to put on a poster and planning to give a brief talk to a room full of statisticians! I winged it pretty successfully and got positive feedback about my work.
The research triangle in NC is an interesting place. There are no stores anywhere around and it definitely has the feel of a place in which you can spend time thinking deeply about many things. I found my short stay there very enjoyable, but I do believe I would become extremely bored there during an extended stay. At one point in my life, I probably would have enjoyed this seclusion more, but I have grown to enjoy the interaction of people and business of life outside of work.
I mistakenly booked my flight for the am instead of the pm which meant that I missed the last day of the conference which was friday. It was too expensive to change my ticket and I actually didn't mind once I realize the talks were not going to be very interesting to me. In fact, I spent most of the conference talk time just working on my laptop and running simulations for my thesis.
I discovered a bug in my many lines of code today. One that was very detrimental to my program and that I had overlooked several time due to it's subtle nature. The bug was so subtle that it was able to mask it self well when I ran my program in simple situation so that everything worked perfectly. It was only when I added "jumps" ty my model, did the but switch into effect.
Most of the problems that I run into actually have nothing to do with the theory I'm thinking being wrong, but rather the limiting power of the computer. For example, the computer might say I am dividing something by zero, which theoretically cant be zero but is so small the computer just says that it is.
Anyways this bug has moved me back into the direction of solving the problem that I had only recently concluded couldnt be solved. That conclusion would have been correct had I not found this bug. I will have to run more tests to be sure again.
This week my department celebrated it's 50 year anniversary. I got to meet up with several of my old colleague that have graduated already and the event was very nice. It's almost like a second family meeting when a group of statisticians are in a room together, there's a bond of jargon, and common purpose. The take raw data which can be very ugly and worthless and make something meaningful, powerful, and beautiful at times.
I missed the entire second day of events as I have been working intently to solve this problem in my office. I am setting a mental clock of two week for this problem; if it is not solved by then I will just give it up.
Tuesday, April 14, 2009
Conclusions
I am rather pleased with the work I achieved today because I believe I have done a sufficient amount of test to conclude that there are two many parameters in my model to obtain exact solutions to my problem. I am fairly confidence in this fact now even though the equations have a very nice form.
I spent half my day dealing with the labcorp people for my pre-screening drug test. They gave me the wrong information and registration number 3 times, wasting 2-3 hours of my day. I am on the 2 page a day thesis plan right now and I will start wrapping up my results by the end of the week.
I am leaving for NC tomorrow and have not done a thing to prepare for my trip!
I spent half my day dealing with the labcorp people for my pre-screening drug test. They gave me the wrong information and registration number 3 times, wasting 2-3 hours of my day. I am on the 2 page a day thesis plan right now and I will start wrapping up my results by the end of the week.
I am leaving for NC tomorrow and have not done a thing to prepare for my trip!
Monday, April 13, 2009
Job cramming
My life has went from 20 to 2000 mph in a couple of days. After the offer, my time line for everything I need to get done has increased exponentially. I am now simultaneously witting my dissertation and trying desperately to finish this last results. I am literally bouncing around the correct answer and the solution continues to elude my grasps. Even when I try to diagnose the issue, I can't pin down my diagnosis with certainty. I am becoming obsessed with solving this problem, as there are several advantages to doing so.
First, it would significantly reduce the amount of material I need to add to my thesis. In summary good results means less writing, however not a good results requires more writing. More writing requires time and time is not something I have much of right now.
I have requested an extension on my start date, hopefully I can get it moved to mid July or August. Currently it is June 15 and I am working under that assumptions. I need to successfully defend my thesis before I leave as well as turn in my thesis document several weeks prior to my defense date.
On top of that I need to find a place and move all my stuff down to S. Florida. I can't even focus all my attention on these thing since I must finish teaching my class as well and I still have a conference to attend in N. Carolina this week. The pressure is on and I am definitely feeling it. I did take friday off to celebrate with my friends, but the rest of the weekend was spent working.
There are many variables I am just counting on to turn out okay. For example, I need to somehow hope that all my advisers and committee will be in town this summer. I am also banking on an extended start date. All these things I cant worry about right now however, what matter the most right now is finishing this freaking problem!
First, it would significantly reduce the amount of material I need to add to my thesis. In summary good results means less writing, however not a good results requires more writing. More writing requires time and time is not something I have much of right now.
I have requested an extension on my start date, hopefully I can get it moved to mid July or August. Currently it is June 15 and I am working under that assumptions. I need to successfully defend my thesis before I leave as well as turn in my thesis document several weeks prior to my defense date.
On top of that I need to find a place and move all my stuff down to S. Florida. I can't even focus all my attention on these thing since I must finish teaching my class as well and I still have a conference to attend in N. Carolina this week. The pressure is on and I am definitely feeling it. I did take friday off to celebrate with my friends, but the rest of the weekend was spent working.
There are many variables I am just counting on to turn out okay. For example, I need to somehow hope that all my advisers and committee will be in town this summer. I am also banking on an extended start date. All these things I cant worry about right now however, what matter the most right now is finishing this freaking problem!
Thursday, April 9, 2009
Job Offer
Long story short: I got the job offer with FPL (Nextera Energy) as a Business Associate in the PMI-Risk Management division.
I was really starting to doubt that I would hear back from them; apparently there was an issue with somebody being sick that had to sign off on the paperwork. The package looks really good and after my euphoria subsides I will need to focus intently on what I need to accomplish to make the transition to S. Florida.
I was really starting to doubt that I would hear back from them; apparently there was an issue with somebody being sick that had to sign off on the paperwork. The package looks really good and after my euphoria subsides I will need to focus intently on what I need to accomplish to make the transition to S. Florida.
Nice Quote
I ran across this quote on facebook and liked it:
"...As if, somehow it makes sense that we can't comprehend our brain's own hundred billion neurons WITH our brain's own hundred billion neurons"
-Albert Goldbarth
"...As if, somehow it makes sense that we can't comprehend our brain's own hundred billion neurons WITH our brain's own hundred billion neurons"
-Albert Goldbarth
Wednesday, April 8, 2009
Quiet Day
Still no call from FPL confirming or denying my status. It is getting to the point where I will have to start assuming that I won't get the position because the last day to get a discount on renewing my lease is tomorrow.
I spent the day debugging my program and found a possible problem. I don't think I am computing the joint expectations correctly; while I don't have a solution yet, this at least narrows the search of my problem. I also discovered several other careless errors in my program. I have re-vised my estimates several times but I believe this should be the last of it.
I spend a decent amount of time examining my students grades in the class; I will spend the first part of tomorrow going over their standing in the class and I might finish my lecture for the semester tomorrow. My students keep asking me when the final is and I keep telling them the university decides that stuff. Truth is, I don't have a clue; I don't care to look at the stuff until the week before finals. The one time I did take the time to alert my class at the beginning of the semester, I accidentally looked up the wrong time. Thankfully a student double checked the correct time and sent me an email.
I went to go see congressman Allen Boyd yesterday; he's a really down to earth guy. Fortunately, few people showed up and I actually got to have a 5 minute conversation with him, it went something like this:
Boyd: He how are you, my name is Allen Boyd?
Me: Yes, I just looked you up on wikipedia
Boyd: O really, what did they have to say about me
Me: That you're a Blue Dog
Boyd: haha...we'll you don't have anything against blue dog's do you?
Me: No, I think that shows a streak of independence, I'm actually a libertarian.
Boyd: Really?
Me: Well I like to say libertarian in theory, democrat in practice
Boyd: I guess George Bush bought alot of you guys to our side huh?
Me: Haha Yea
Boyd: Well sound like you might be a Blue Dog too!
He spoke about the bailout for about half and hour and opened the floor for questions. I really like his answers on guns and healthcare.
One guns he basically said that the only time some people who live in miami and new york see a gun is when it is used in a crime. In rural areas guns are mostly used for hunting which is why there is always this divide on this subject.
On healthcare, he actually gave some good arguments for not rushing to a single payer system. Since all doctors would essentially be paid and regulated by the government, there are bound to be inefficiencies created.
I'm glad I went, I see why it's important for politicians to go out and mingle with the people; He seems like a decent and thought guy and I'd probably vote for him next go around.
I spent the day debugging my program and found a possible problem. I don't think I am computing the joint expectations correctly; while I don't have a solution yet, this at least narrows the search of my problem. I also discovered several other careless errors in my program. I have re-vised my estimates several times but I believe this should be the last of it.
I spend a decent amount of time examining my students grades in the class; I will spend the first part of tomorrow going over their standing in the class and I might finish my lecture for the semester tomorrow. My students keep asking me when the final is and I keep telling them the university decides that stuff. Truth is, I don't have a clue; I don't care to look at the stuff until the week before finals. The one time I did take the time to alert my class at the beginning of the semester, I accidentally looked up the wrong time. Thankfully a student double checked the correct time and sent me an email.
I went to go see congressman Allen Boyd yesterday; he's a really down to earth guy. Fortunately, few people showed up and I actually got to have a 5 minute conversation with him, it went something like this:
Boyd: He how are you, my name is Allen Boyd?
Me: Yes, I just looked you up on wikipedia
Boyd: O really, what did they have to say about me
Me: That you're a Blue Dog
Boyd: haha...we'll you don't have anything against blue dog's do you?
Me: No, I think that shows a streak of independence, I'm actually a libertarian.
Boyd: Really?
Me: Well I like to say libertarian in theory, democrat in practice
Boyd: I guess George Bush bought alot of you guys to our side huh?
Me: Haha Yea
Boyd: Well sound like you might be a Blue Dog too!
He spoke about the bailout for about half and hour and opened the floor for questions. I really like his answers on guns and healthcare.
One guns he basically said that the only time some people who live in miami and new york see a gun is when it is used in a crime. In rural areas guns are mostly used for hunting which is why there is always this divide on this subject.
On healthcare, he actually gave some good arguments for not rushing to a single payer system. Since all doctors would essentially be paid and regulated by the government, there are bound to be inefficiencies created.
I'm glad I went, I see why it's important for politicians to go out and mingle with the people; He seems like a decent and thought guy and I'd probably vote for him next go around.
Tuesday, April 7, 2009
Still haven't found the way
I was slightly annoyed with the computational results I've been getting because it is hard to diagnose whether my program just isn't working or I don't have enough data for it to work. If I could easily distinguish which one it is it would make life simpler.
Things I know:
1. If I know Y and X, I can estimate Z
2. If I know Y and Z, I can estimate X
3. If I only know Y, I can combines (1) and (2) to estimate X and Z in simple problems
I still haven't figured out how to estimate X and Z by combining 1 and 2 in a slightly more complicated problem.
I feel as if all the pieces to the puzzle are there, I just need to make the final leap. Still on word from FPL, so I spent the day in suspense which was actually frustrating. I just want to know is it a yes or no. Good thing is that I finally got my check from them yesterday for my expenses that I incurred while traveling down there to interview.
I also got a letter from the ACT saying I didn't get their internship; I really didn't care about that one and wasn't looking forward to IOWA anyhow. I would have reject me probably. I submitted my paperwork for the IMSM 2009 workshop and was actually please with my essay since I had it reviewed by someone more talented with words than myself.
US rep Allen Boyd ( I think that's his name) will be on campus today at 7. Not really sure what he's talking about but I think I would like to attend.
Things I know:
1. If I know Y and X, I can estimate Z
2. If I know Y and Z, I can estimate X
3. If I only know Y, I can combines (1) and (2) to estimate X and Z in simple problems
I still haven't figured out how to estimate X and Z by combining 1 and 2 in a slightly more complicated problem.
I feel as if all the pieces to the puzzle are there, I just need to make the final leap. Still on word from FPL, so I spent the day in suspense which was actually frustrating. I just want to know is it a yes or no. Good thing is that I finally got my check from them yesterday for my expenses that I incurred while traveling down there to interview.
I also got a letter from the ACT saying I didn't get their internship; I really didn't care about that one and wasn't looking forward to IOWA anyhow. I would have reject me probably. I submitted my paperwork for the IMSM 2009 workshop and was actually please with my essay since I had it reviewed by someone more talented with words than myself.
US rep Allen Boyd ( I think that's his name) will be on campus today at 7. Not really sure what he's talking about but I think I would like to attend.
Monday, April 6, 2009
New Start
It was good that I waited to speak with my adviser before I continued working on the problem. He pointed out a couple of things that made me realize that I had overlooked several details in my programs. Basically I wouldn't have gotten any correct results until I had dealt with these issues. I spent most of the day correcting these problems and preparing my notes for tomorrows lecture.
I got a call from FPL today informing me that they should have a decision for me within a couple of days. The suspense is killing me haha. I am actually quite content either way right now. I wouldn't mind getting the job, it would be an exit out of graduate school and I could start wrapping things up quickly assuming I get this problem worked out. I think it's a great location and opportunity.
I also think that there will be several opportunities for me even without the job. There is JSM 2009 in August which will no doubt have multiple head hunters. Also the summer workshop assuming I get in would be a good look. There are all several consulting opportunities that could pursuing this summer.
Basically I strongly believe that God has already made the decision for me. Whatever it is, I believe that He know where I should be; it's my job to understand why. Looking back in my life there have been so many decisions that have caused dramatic changes in my life. This will no doubt be one of them, as big as my decision to move to FSU four years ago.
Life has been going pretty smoothly and I'm looking forward to going on a run and hitting the gym in about an hour.
I got a call from FPL today informing me that they should have a decision for me within a couple of days. The suspense is killing me haha. I am actually quite content either way right now. I wouldn't mind getting the job, it would be an exit out of graduate school and I could start wrapping things up quickly assuming I get this problem worked out. I think it's a great location and opportunity.
I also think that there will be several opportunities for me even without the job. There is JSM 2009 in August which will no doubt have multiple head hunters. Also the summer workshop assuming I get in would be a good look. There are all several consulting opportunities that could pursuing this summer.
Basically I strongly believe that God has already made the decision for me. Whatever it is, I believe that He know where I should be; it's my job to understand why. Looking back in my life there have been so many decisions that have caused dramatic changes in my life. This will no doubt be one of them, as big as my decision to move to FSU four years ago.
Life has been going pretty smoothly and I'm looking forward to going on a run and hitting the gym in about an hour.
Sunday, April 5, 2009
Enjoyable Weekend
This weekends was one of the better weekends that I can remember me having in Tallahassee in quite some time. It is sunday afternnon and I haven't done any research, nor prepared for my class next week; and I still don't feel rushed to do so.
One great thing about my schedule this semester is that monday is practically a free day for me to get a manageable amount of work done. And so while I wouldn't mind getting a few things done this weekend, I'm sure I can push it all to monday.
I ended up going out on Friday with a couple of friends; and had a pretty enjoyable time, which is something that doesn't happen often. As I get older, I find the nightlife in Tallahassee to be a much younger crowd than I prefer to hang around; plus I have a preference for spots that play alot of reggae music and that just isn't a popular form here. Fortunately there was about 30-45 minutes played on the night we went out.
I went to the department picnic saturday around noon; due to the lack of sleep I had form the night before and a game of soccer at Tom Brown park I was very tired when I got back. I practically slept the rest of the afternnoon. I am still in bed right now; I only plan to do work which I can do on my laptop anyhow; there are couple of thing I would like to attempt to do at some point today:
1. Start thinking about a revised version of my student's final project. I already have a template made, I just want to structure it a bit more.
2. Finish the essay part of my application that is due on April 15. There is a weeklong program this summer for graduate students to work on a team solving industrial problems for several companies. I think it would be an excellent way to get a foot in the door for possible employment if FPL doesnt work out; which I should here something about this week.
One reason I am not attempting new things with my research is that I feel as if I can gone quite a while without getting feedback from my advisor. Often when I do this I end up missing crucial things I did not see before which would have saved me a lot of meaningless pursuits. So I am being patient and will wait until tomorrow to discuss my progress and remaining issues.
One great thing about my schedule this semester is that monday is practically a free day for me to get a manageable amount of work done. And so while I wouldn't mind getting a few things done this weekend, I'm sure I can push it all to monday.
I ended up going out on Friday with a couple of friends; and had a pretty enjoyable time, which is something that doesn't happen often. As I get older, I find the nightlife in Tallahassee to be a much younger crowd than I prefer to hang around; plus I have a preference for spots that play alot of reggae music and that just isn't a popular form here. Fortunately there was about 30-45 minutes played on the night we went out.
I went to the department picnic saturday around noon; due to the lack of sleep I had form the night before and a game of soccer at Tom Brown park I was very tired when I got back. I practically slept the rest of the afternnoon. I am still in bed right now; I only plan to do work which I can do on my laptop anyhow; there are couple of thing I would like to attempt to do at some point today:
1. Start thinking about a revised version of my student's final project. I already have a template made, I just want to structure it a bit more.
2. Finish the essay part of my application that is due on April 15. There is a weeklong program this summer for graduate students to work on a team solving industrial problems for several companies. I think it would be an excellent way to get a foot in the door for possible employment if FPL doesnt work out; which I should here something about this week.
One reason I am not attempting new things with my research is that I feel as if I can gone quite a while without getting feedback from my advisor. Often when I do this I end up missing crucial things I did not see before which would have saved me a lot of meaningless pursuits. So I am being patient and will wait until tomorrow to discuss my progress and remaining issues.
Friday, April 3, 2009
Flawed Logic
My results yesterday was actually premised on the knowledge of 2 parameters in the system. While this is significant project given the amount of parameters in the model; I must still work on this problem to element the need for this knowledge.
My model is so complex now that I have to decide between:
1. the time it's takes to get results
2. the accuracy of my results
If I increase my data observation I get better results, however it could take hours longer to finish. When you are certain that what you are doing is correct this is fine. However when you are just testing out models it is not cool to wait 2 hours only to find out you accidentally added a line of code that wasn't supposed to be there, which mean you must run the whole thing again!
Write now I am running a very computational intensive step known as a forward backward smoother. Everything else in my program runs quickly, however it takes 6-7 minutes to run this step, and I need to run it at least 10 times. So I try to time my running of the program around lunch time and when I teach. I really should get in the habit of running the really long simulations over night when I am sleeping.
I am sure other statisticians in the department would scoff at my simulations only taking hours as I know some of them are running simulations that take months and in some cases years running on 4-5 different computers! In our computer lab it is not uncommon to see multiple computers with a "please don't touch" paper over it.
Once again, I am giving a brief 2-3 minute talk about my research to a visiting professor at 4. Other than that I just plan on playing around with my results until next week when I can meet with my adviser. I feel as I am making progress at a reasonable pace right now so I am not anxious or anything. This is a very rare feeling in my life.
My model is so complex now that I have to decide between:
1. the time it's takes to get results
2. the accuracy of my results
If I increase my data observation I get better results, however it could take hours longer to finish. When you are certain that what you are doing is correct this is fine. However when you are just testing out models it is not cool to wait 2 hours only to find out you accidentally added a line of code that wasn't supposed to be there, which mean you must run the whole thing again!
Write now I am running a very computational intensive step known as a forward backward smoother. Everything else in my program runs quickly, however it takes 6-7 minutes to run this step, and I need to run it at least 10 times. So I try to time my running of the program around lunch time and when I teach. I really should get in the habit of running the really long simulations over night when I am sleeping.
I am sure other statisticians in the department would scoff at my simulations only taking hours as I know some of them are running simulations that take months and in some cases years running on 4-5 different computers! In our computer lab it is not uncommon to see multiple computers with a "please don't touch" paper over it.
Once again, I am giving a brief 2-3 minute talk about my research to a visiting professor at 4. Other than that I just plan on playing around with my results until next week when I can meet with my adviser. I feel as I am making progress at a reasonable pace right now so I am not anxious or anything. This is a very rare feeling in my life.
Thursday, April 2, 2009
One of those days...
I had one of those day in which I had extreme satisfaction with what I do. Partly fueled by the fact that I have come up with a way to solve the problem I have been working on. Right now the results are very crude and the theory very messy and disorganized, but I'm am dancing around a refined solution and I am quite confident in my results.
It's very interesting how research works, I remember when I first stated the problem and it seem almost impossible but something to just play around with. After playing with the problem, I started to see some possible solutions. By working through many possible avenues, I stumbled across a method that appears to spit out what I've been looking for. When I am closing in on a solution, I feel as if my mind works faster and faster as if I started accelerating down a hill. It will be interesting to look back at my entries to track all the ups and down of my thought process when I finally converge to a solution
I had very small classes today probably due to the rain, but I felt my lectures went well; I prefer teaching smaller sections anyway. I had to leave my office early last night so they could finish installing the new carpet in the department, they finally finished today and it will be nice for everything to be back to normal. Due to all the rain last week, the department picnic has been moved to this coming Saturday; I hope it is beautiful day because I am looking forward to going to the rez.
It's very interesting how research works, I remember when I first stated the problem and it seem almost impossible but something to just play around with. After playing with the problem, I started to see some possible solutions. By working through many possible avenues, I stumbled across a method that appears to spit out what I've been looking for. When I am closing in on a solution, I feel as if my mind works faster and faster as if I started accelerating down a hill. It will be interesting to look back at my entries to track all the ups and down of my thought process when I finally converge to a solution
I had very small classes today probably due to the rain, but I felt my lectures went well; I prefer teaching smaller sections anyway. I had to leave my office early last night so they could finish installing the new carpet in the department, they finally finished today and it will be nice for everything to be back to normal. Due to all the rain last week, the department picnic has been moved to this coming Saturday; I hope it is beautiful day because I am looking forward to going to the rez.
Wednesday, April 1, 2009
Significant Progress
I made significant progress on my problem today, instead of re-specifying my model I rescaled the data so I would get extremely large numbers. I have completed the verification process of my parameter estimates using the first and simple test.
The last remaining step is the verify the parameter estimates under the second test which is a more complicated situation that I don't feel like tackling today. Actually I do and I don't; I am curious to see if this results works but I have a ton of code that I will have to modify to do so and I don't think I have it in me today.
I still need to prepare for my lecture tomorrow. My students took their exam yesterday and I don't think many did that well. One girl decided to drop and several other are worried about their standing in the class. I think most of them are causing themselves more anxiety than is warranted being that most people that take my class makes A's, but I'd rather them concerned than complacent.
I wrote a recommendation letter for a student two years ago whom I could barely remember. The only reason I did so I guess was because so few students ask me to do so, but if I was swamped with requests I doubt I would have done it. The way she approached me probably made me less enthusiastic to do so.
I went to a talk yesterday where some Christian philosopher came to campus to argue seven different 'proofs' of the existence of God. It was almost to esoteric in nature and I could tell most people in the room did not understand his logic which went something like this:
1. If God exists, then objective moral values exist
2. Objective moral values exist
3. Therefore God exists
I prefer the ontological proof of the existences of God, however the version he gave had 7 or 8 logical sequences which I thought was a bit much. My version is much simpler:
1. We can conceive of a God by definition to be the greatest thing that can be thought of, if there were anything greater than that would be God
2. It is greater for such a God to exist than not to exist; since not existing would not be the greatest thing that could be thought of.
3. Therefore God exists
Such logic depends on the assumption that existence is greater than non-existence which from our perspective seems at least logical.
He did mention a nice example about how infinite-infinite=infinite, or 3, or 5; that was probably the only thing he really explained clearly.
I traded my old nonworking laptop for a fancy working photographer's camera, I have not use for it but will give it to my sister when I go back home. I spoke with FPL yesterday and was told there still was not a decision on the position yet, however they assured me that should know something by April 10. I really would like to know how my summer is going to be ASAP so hopefully I hear something by the end of this week or early next week.
The last remaining step is the verify the parameter estimates under the second test which is a more complicated situation that I don't feel like tackling today. Actually I do and I don't; I am curious to see if this results works but I have a ton of code that I will have to modify to do so and I don't think I have it in me today.
I still need to prepare for my lecture tomorrow. My students took their exam yesterday and I don't think many did that well. One girl decided to drop and several other are worried about their standing in the class. I think most of them are causing themselves more anxiety than is warranted being that most people that take my class makes A's, but I'd rather them concerned than complacent.
I wrote a recommendation letter for a student two years ago whom I could barely remember. The only reason I did so I guess was because so few students ask me to do so, but if I was swamped with requests I doubt I would have done it. The way she approached me probably made me less enthusiastic to do so.
I went to a talk yesterday where some Christian philosopher came to campus to argue seven different 'proofs' of the existence of God. It was almost to esoteric in nature and I could tell most people in the room did not understand his logic which went something like this:
1. If God exists, then objective moral values exist
2. Objective moral values exist
3. Therefore God exists
I prefer the ontological proof of the existences of God, however the version he gave had 7 or 8 logical sequences which I thought was a bit much. My version is much simpler:
1. We can conceive of a God by definition to be the greatest thing that can be thought of, if there were anything greater than that would be God
2. It is greater for such a God to exist than not to exist; since not existing would not be the greatest thing that could be thought of.
3. Therefore God exists
Such logic depends on the assumption that existence is greater than non-existence which from our perspective seems at least logical.
He did mention a nice example about how infinite-infinite=infinite, or 3, or 5; that was probably the only thing he really explained clearly.
I traded my old nonworking laptop for a fancy working photographer's camera, I have not use for it but will give it to my sister when I go back home. I spoke with FPL yesterday and was told there still was not a decision on the position yet, however they assured me that should know something by April 10. I really would like to know how my summer is going to be ASAP so hopefully I hear something by the end of this week or early next week.
Subscribe to:
Posts (Atom)