Unemployment In The U.S. Will Fall To 6.2% By 2014 [View article]
How many democrats are needed to bias unemployment figures?
I do not consider any possibility that the Current Population Survey conducted by the U.S. Census Bureau for September 2012 is biased by CB or by the BLS, This is not the case. There is another hypothetical way to bias the data. There are around 70000 households surveyed by the CB. These households include approximately 200000 persons (mean household is 2.5 people). All these people (excluding several percent not responding ones) answer a few questions associated with their current status: employed, unemployed or not in the labor force. There current level of civilian labor force is approximately 155,000,00 with 12,000,000 unemployed. These figures are calculated by a projection of 200,000 to 310,000,000 using population controls. In essence, one person represents 1550 people.
How many people are needed to increase the rate of unemployment by 0.1%? The rate of unemployment is calculated as the ratio of the number of unemployed and labor force. So, 0.1% of unemployment rate with the level of labor force of 155,000,000 corresponds to 155000. Since one person in the CPS represents 1550 people, one needs only 100 people to increase the rate of unemployment by 0.1%. To decrease the rate by 0.3% , only 300 (democrats -Spartans?) are needed.
Income Inequality For Households: A Long Biased History Of Gini And Mean Income [View article]
It is difficult to judge without quality data. I was trying to play a bit around this problem (http://bit.ly/T2MmFs) but data for high income households of all sizes are not measured accurately enough. This is a question why the Census does not attack this important social/cultural/policy... problem appropriately? I have to notice again that the personal incomes obtained from the very same households do not show any fall in Gini ratio since 1994. This is a fact reported by the CB. I have calculated them back in the past to 1967 ( http://bit.ly/QJhEnW) and found no increase as well.
Income Inequality For Households: A Long Biased History Of Gini And Mean Income [View article]
I would not agree with the assumption that poor households have a higher propensity to split. Obviously, poor families save lots of money living together. I would suggest that rich families have more reasons to break - their pieces have higher economic and financial viability. And the probability to break should increase with the average income. Unfortunately, the Census Bureau does not provide these important measurements for a longer time period and we can not prove any of the statements quantitatively. As you know, the CPS covers only 70,000 households and the higher incomes are not well represented. Also the CB limited the upper income bound to only $100,000 for all household sizes ($200,000 since 2009). For your convenience, I have more on household and personal income inequality in my blog (link is available at my author page).
Daily State Of The Markets: What Are The Cycles Saying Now? [View article]
I wrote couple days ago that the level of S&P 500 is highly overvalued. Sold SPY this morning to fix 10% profit. Turbulent time is ahead. The S&P 500 will likely fall to 1375 in October/November. An opportunity to get another 10% by the end of 2013.
I also have to admit that the model based on the number of 9-year-olds does not work well. The 2010 census has shown why. The Census Bureau severely suppresses actual changes in population cohorts and no true changes can be recovered. Therefore, the model failure is partly due to artificiality of population data.
Deflation is coming - oil falls and drives prices down. Formal deflation in 2012 will be clear in the GDP deflator Real economic growth is suppressed - might be around 1%. S&P 500 will be slowly growing to 1500 by the end of 2013. It does not grow much in 2012.
This is an article about HAL http://seekingalpha.co... I'll write about SLB as well. Actually, I have an idea compare all energy companies from S&P 500 in a larger article, likely more academic and published in the journal where Im an editor.
Predicting The Future Of The Oracle Of Omaha [View article]
The model has to be the same at least 6 to 12 months.When one or both defining CPIs change the previous model fails. The model also allows for time delays from 0 to 12 months in both directions. I am not aware about any logical underpinning of share prices which works. The evolution of share prices is usually considered as stochastic, i.e. these are pure innovations without any deterministics explanation. I assume that there is a CPI (might not be perfect) which contains the good and services influencing the share price. Then this CPI correlates with the share price more than anything else. Such CPIs do not exist in many cases. For some companies, the current CPI set does contain two CPIs which define the share during several months. For these companies I report models. At the same time I do not know how CPIs are converted into share prices. Investors might look at Dairy Queen or GEICOand judge on the overall performance of BRK.B (i.e. on good or poor management decisions). Thus , even a small change in their performance can change the share price.
Predicting The Future Of The Oracle Of Omaha [View article]
1. I do follow previously reported models (see my articles on COP). I report only those models which are stable during several months. So, they always has a history of several months. SOme models stay years (HPQ). 2. The model where CPIs drive a share price is rather physics - prices drives price. For example, CPI of energy drives COP. Nobody denies this possibility. 3. BRK,B just a hypothesis as any other model. It works where it works. Before it stops working it's worth something.
Apache Looks To Be A Better Investment Than ConocoPhillips [View article]
I do use monthly closing prices and monthly CPI estimates. The procedure is based on linear regression of around 100 months for the advanced model (with time lags) and just on visual fit for the original (CC and C) model. I did not calculate t-stat but checked the residual error for unit roots (i.e. cointegration between observed and predicted prices) and other simple stat features. I have no doubt that t-stat is good since the predicted price is very sensitive to the coefficients. In any case, it is easy to estimate for any given model.
I did calculated and presented some returns for some reliable deterministic models in my book - http://amzn.to/wmO6rS . However, returns was not teh main topic.
Apache Looks To Be A Better Investment Than ConocoPhillips [View article]
Right. I have to disclose the purchase but this is a part of a diversified portfolio not a COP share purchase. I am going to enter the stock market a little later in April, when the S&P 500 is at 1350. Then I consider APA and BTU as very attractive ones. At the same time, this is kind of technical analysis without any prejudice. You may judge what's it worth. It has beeb working well for COP since 2003.
Why Peabody Energy Is Highly Undervalued [View article]
Thank you for this comment. Did both disruptive impacts start in April 2011? The SA editor has removed the question mark from my original title of thsi article. Actually, I wonder what was the reason behind the deviation?
Why Peabody Energy Is Highly Undervalued [View article]
Right. This is my first investment - should be defensive and a part of a big portfolio. GOing to enter stocks (likely BTU and NFX) directly when S&P 500 is at 1350 - expect this level by May.
Unemployment In The U.S. Will Fall To 6.2% By 2014 [View article]
I do not consider any possibility that the Current Population Survey conducted by the U.S. Census Bureau for September 2012 is biased by CB or by the BLS, This is not the case. There is another hypothetical way to bias the data. There are around 70000 households surveyed by the CB. These households include approximately 200000 persons (mean household is 2.5 people). All these people (excluding several percent not responding ones) answer a few questions associated with their current status: employed, unemployed or not in the labor force. There current level of civilian labor force is approximately 155,000,00 with 12,000,000 unemployed. These figures are calculated by a projection of 200,000 to 310,000,000 using population controls. In essence, one person represents 1550 people.
How many people are needed to increase the rate of unemployment by 0.1%? The rate of unemployment is calculated as the ratio of the number of unemployed and labor force. So, 0.1% of unemployment rate with the level of labor force of 155,000,000 corresponds to 155000. Since one person in the CPS represents 1550 people, one needs only 100 people to increase the rate of unemployment by 0.1%. To decrease the rate by 0.3% , only 300 (democrats -Spartans?) are needed.
Crude And Steel Prices Still In Sync [View article]
currently, only a blog - http://bit.ly/VlJBVV
We build up a site StockMarketScience to be launched any time soon.
Income Inequality For Households: A Long Biased History Of Gini And Mean Income [View article]
I have to notice again that the personal incomes obtained from the very same households do not show any fall in Gini ratio since 1994. This is a fact reported by the CB. I have calculated them back in the past to 1967 ( http://bit.ly/QJhEnW) and found no increase as well.
Income Inequality For Households: A Long Biased History Of Gini And Mean Income [View article]
Unfortunately, the Census Bureau does not provide these important measurements for a longer time period and we can not prove any of the statements quantitatively. As you know, the CPS covers only 70,000 households and the higher incomes are not well represented. Also the CB limited the upper income bound to only $100,000 for all household sizes ($200,000 since 2009).
For your convenience, I have more on household and personal income inequality in my blog (link is available at my author page).
Daily State Of The Markets: What Are The Cycles Saying Now? [View article]
Time To Buy SPY [View instapost]
Time To Buy SPY [View instapost]
Real economic growth is suppressed - might be around 1%.
S&P 500 will be slowly growing to 1500 by the end of 2013. It does not grow much in 2012.
Baker Hughes Is Likely Undervalued [View article]
I'll write about SLB as well. Actually, I have an idea compare all energy companies from S&P 500 in a larger article, likely more academic and published in the journal where Im an editor.
Predicting The Future Of The Oracle Of Omaha [View article]
I am not aware about any logical underpinning of share prices which works. The evolution of share prices is usually considered as stochastic, i.e. these are pure innovations without any deterministics explanation.
I assume that there is a CPI (might not be perfect) which contains the good and services influencing the share price. Then this CPI correlates with the share price more than anything else. Such CPIs do not exist in many cases. For some companies, the current CPI set does contain two CPIs which define the share during several months. For these companies I report models.
At the same time I do not know how CPIs are converted into share prices. Investors might look at Dairy Queen or GEICOand judge on the overall performance of BRK.B (i.e. on good or poor management decisions). Thus , even a small change in their performance can change the share price.
Predicting The Future Of The Oracle Of Omaha [View article]
2. The model where CPIs drive a share price is rather physics - prices drives price. For example, CPI of energy drives COP. Nobody denies this possibility.
3. BRK,B just a hypothesis as any other model. It works where it works. Before it stops working it's worth something.
Apache Looks To Be A Better Investment Than ConocoPhillips [View article]
Apache Looks To Be A Better Investment Than ConocoPhillips [View article]
I did not calculate t-stat but checked the residual error for unit roots (i.e. cointegration between observed and predicted prices) and other simple stat features. I have no doubt that t-stat is good since the predicted price is very sensitive to the coefficients. In any case, it is easy to estimate for any given model.
I did calculated and presented some returns for some reliable deterministic models in my book - http://amzn.to/wmO6rS . However, returns was not teh main topic.
Apache Looks To Be A Better Investment Than ConocoPhillips [View article]
At the same time, this is kind of technical analysis without any prejudice. You may judge what's it worth. It has beeb working well for COP since 2003.
Why Peabody Energy Is Highly Undervalued [View article]
Why Peabody Energy Is Highly Undervalued [View article]