My last day of work at IBM was just over 5 years ago. So I was bit surprised to see in my mail today a check from the IBM payroll department. It turned out that on January 28, 2014 the US patent office had issued patent 8,639,737 for an invention I made while employed by IBM which entitled to me to a small award from IBM's patent incentive program. The application was filed March 8, 2008 and was derived from an earlier application filed May 27, 2005. Which all seems like a very long time ago at this point.
While it was nice in a way to get this reminder of a past life it is also going to complicate my taxes for 2014. This is New York state income so I will have to file a New York state return and I will likely end up over withheld with respect to social security.
Friday, February 28, 2014
Monday, February 17, 2014
Priors
Recently Dylan Farrow, the adopted daughter of Woody Allen and Mia Farrow, repeated allegations that Allen had sexually molested her when she was 7 years old. These allegations were first made in 1982 during Allen and Farrow's acrimonious breakup (triggered by Farrow discovering Allen was involved with her adopted daughter Soon-Yi). The allegations were investigated at the time and no criminal charges were pressed against Allen although the judge in a custody battle between Allen and Farrow seems to have given them some credence.
The revived charges led to considerable internet discussion with some people quite convinced of Allen's guilt while others were far more skeptical. I think it is worth noting that this disagreement can reflect different prior beliefs about how common sexual abuse of children is rather than different opinions about the strength of the evidence in Allen's case.
Suppose for example that some people believe that fathers rarely molest their children say .1% of the time while others believe it is fairly common say 10% of the time. Suppose further that a credible accusation of abuse will be made in 1% of the cases where no abuse occurred and (for simplicity) in 100% of the cases where abuse did occur. Then what fraction of credible accusations will in fact be true?
If sexual abuse is rare then out of every million men there will be 1000 cases of actual abuse (hence 1000 true accusations) and 9990 false accusations (.01 * 999000). So only 9.1% of the credible accusations will actually be true.
On the other hand if abuse is common there will be 100000 true accusations and 9000 false accusations. So in this case 91.7% of the accusations will be true. Hence in a case like Allen's your prior beliefs can determine whether you see the evidence of guilt as weak or strong.
Of course I made up the above numbers to illustrate my point. The reader can try different numbers and see how the fraction of false accusations changes. I will note that assuming (more realistically) that less than 100% of actual abuse is reported leads to an increase in the fraction of false accusations.
This false positive problem also occurs in other contexts. For example if a medical condition is rare a positive test result may be more likely test error than a true positive even if the test is fairly reliable.
The revived charges led to considerable internet discussion with some people quite convinced of Allen's guilt while others were far more skeptical. I think it is worth noting that this disagreement can reflect different prior beliefs about how common sexual abuse of children is rather than different opinions about the strength of the evidence in Allen's case.
Suppose for example that some people believe that fathers rarely molest their children say .1% of the time while others believe it is fairly common say 10% of the time. Suppose further that a credible accusation of abuse will be made in 1% of the cases where no abuse occurred and (for simplicity) in 100% of the cases where abuse did occur. Then what fraction of credible accusations will in fact be true?
If sexual abuse is rare then out of every million men there will be 1000 cases of actual abuse (hence 1000 true accusations) and 9990 false accusations (.01 * 999000). So only 9.1% of the credible accusations will actually be true.
On the other hand if abuse is common there will be 100000 true accusations and 9000 false accusations. So in this case 91.7% of the accusations will be true. Hence in a case like Allen's your prior beliefs can determine whether you see the evidence of guilt as weak or strong.
Of course I made up the above numbers to illustrate my point. The reader can try different numbers and see how the fraction of false accusations changes. I will note that assuming (more realistically) that less than 100% of actual abuse is reported leads to an increase in the fraction of false accusations.
This false positive problem also occurs in other contexts. For example if a medical condition is rare a positive test result may be more likely test error than a true positive even if the test is fairly reliable.
Wednesday, February 5, 2014
Practice to Deceive
I recently read "Practice to Deceive", a 2013 account by Ann Rule of a killing in Washington state and the eventual trial and conviction of the killer. Long ago I read Rule's 1983 novel, "Possession", and really liked it. However this seems to have been a minority opinion since the rest of her many books (as far as I know) have been in the "true crime" genre. I have read a few of these and didn't like them as much. She is competent enough at what she does but the books will mostly appeal to fans of the genre (which I am only in a very minor way).
I found this book pretty mediocre. I just didn't find the crime or the investigation or the cast of characters all that interesting. Another issue is that no clear cut motive was ever established (although there seems little doubt of the convicted killer's guilt). This is a little unsatisfying if you like things wrapped up neatly at the end (although perhaps some readers will enjoy speculating about what the full story was).
So in my view there is little reason to read this book unless you are a huge fan of the genre (or Ann Rule).
I found this book pretty mediocre. I just didn't find the crime or the investigation or the cast of characters all that interesting. Another issue is that no clear cut motive was ever established (although there seems little doubt of the convicted killer's guilt). This is a little unsatisfying if you like things wrapped up neatly at the end (although perhaps some readers will enjoy speculating about what the full story was).
So in my view there is little reason to read this book unless you are a huge fan of the genre (or Ann Rule).
Monday, February 3, 2014
The Elements of Investing
I recently read "The Elements of Investing" a short book of elementary investment advice by Burton Malkiel and Charles D. Ellis. I am a bit conflicted about this book. On the one hand I agree with almost everything in it. But on the other hand I didn't find it very interesting because it was largely making arguments I already knew for positions I already agreed with. And I am not sure how convincing it would be if you weren't already in agreement.
Malkiel and Ellis start by making the sensible point that it doesn't matter how good an investor you are if you don't have anything to invest. So for most people (starting without sizable inherited wealth) accumulating substantial wealth will require saving a lot of their income. This may seem obvious but in my view is sometimes neglected. Realistically saving more of your income is a surer path to increasing your wealth than attempting to achieve above average investment returns.
Malkiel and Ellis emphasize the importance of starting to save when you are young to allow more time for investment returns to compound. There is some truth to this but the effect can be exaggerated. In many cases your income will grow as you get older allowing you to put aside increasing amounts each year making your early years of saving even compounded less significant to the final result than if you were putting aside the same amount every year as the illustrative examples (including the one in this book) often assume.
Another point of disagreement is that Malkiel and Ellis uncritically support borrowing to buy a house. This can be a good idea especially if you have a stable employment situation but has real risks which the authors ignore.
Malkiel and Ellis go on to discuss how people should invest their accumulating savings. They conclude that a mix of low cost index funds is the way to go. This is simple and easy to implement and difficult to beat even for full time professional investors. I agree with this in theory (although I have deviated from this in practice mostly to my cost).
So to sum up this book makes the case that most people should invest largely in index funds. I don't really disagree but found the book a bit boring. So if you aren't familiar with the case for index funds this book may be an useful introduction but if you are I don't think this book adds a lot.
Malkiel and Ellis start by making the sensible point that it doesn't matter how good an investor you are if you don't have anything to invest. So for most people (starting without sizable inherited wealth) accumulating substantial wealth will require saving a lot of their income. This may seem obvious but in my view is sometimes neglected. Realistically saving more of your income is a surer path to increasing your wealth than attempting to achieve above average investment returns.
Malkiel and Ellis emphasize the importance of starting to save when you are young to allow more time for investment returns to compound. There is some truth to this but the effect can be exaggerated. In many cases your income will grow as you get older allowing you to put aside increasing amounts each year making your early years of saving even compounded less significant to the final result than if you were putting aside the same amount every year as the illustrative examples (including the one in this book) often assume.
Another point of disagreement is that Malkiel and Ellis uncritically support borrowing to buy a house. This can be a good idea especially if you have a stable employment situation but has real risks which the authors ignore.
Malkiel and Ellis go on to discuss how people should invest their accumulating savings. They conclude that a mix of low cost index funds is the way to go. This is simple and easy to implement and difficult to beat even for full time professional investors. I agree with this in theory (although I have deviated from this in practice mostly to my cost).
So to sum up this book makes the case that most people should invest largely in index funds. I don't really disagree but found the book a bit boring. So if you aren't familiar with the case for index funds this book may be an useful introduction but if you are I don't think this book adds a lot.
Sunday, February 2, 2014
Utility Rates
When I moved from Ossining NY to Princeton NJ I didn't notice much change in my utility (gas and electric) bill. This was a bit of a pleasant surprise as my new townhouse is considerably bigger and is also an end unit with much more exterior wall exposure (compared to the middle unit I had in Ossining). I had some vague thoughts that perhaps my new unit was better insulated or had more efficient heating and cooling systems but upon investigation it appears the main reason is that utility rates are much lower in Princeton than in Ossining.
In 2012 I paid PSEG (my Princeton gas and electric utility) $998 for 890 Ccf of natural gas while in 2011 I paid ConEd (my Ossining utility) $1090 for 562 Ccf. So my average rate dropped from $1.94 per Ccf to $1.12 per Ccf. Similarly my PSEG electric bill in 2012 was $947 for 5193 kWh while in 2011 ConEd charged me $890 for 3451 kWh. So my average rate dropped from $.258 per kWh to $.182 per kWh.
I was aware that New York utility rates are high compared to the national average but the difference with New Jersey is more than I expected. For context the national average natural gas rate seems to be about $1.10 per Ccf and the national average electric rate is about $.12 per kWh (in both cases for residential users).
In 2012 I paid PSEG (my Princeton gas and electric utility) $998 for 890 Ccf of natural gas while in 2011 I paid ConEd (my Ossining utility) $1090 for 562 Ccf. So my average rate dropped from $1.94 per Ccf to $1.12 per Ccf. Similarly my PSEG electric bill in 2012 was $947 for 5193 kWh while in 2011 ConEd charged me $890 for 3451 kWh. So my average rate dropped from $.258 per kWh to $.182 per kWh.
I was aware that New York utility rates are high compared to the national average but the difference with New Jersey is more than I expected. For context the national average natural gas rate seems to be about $1.10 per Ccf and the national average electric rate is about $.12 per kWh (in both cases for residential users).
Subscribe to:
Posts (Atom)
