Big Data. Bureau of Labor Statistics. Survey data. Employment Big Data.  Those are all things that calculating worklife expectancy for U.S. workers requires.  Worklife expectancy is similar to life expectancy and indicates how long a person can be expected to be active in the workforce over their working life.  The worklife expectancy figure takes into account the anticipated to time out of the market due to unemployment, voluntary leaves, attrition, etc.

The goal of our recent work is to update the Millimet et al (2002) worklife expectancy paper and account for more recent CPS data. Their paper uses data from  the 1992 to 2000 time period. Our goal is to update that paper using data from 2000 to 2013 and  see if estimating the Millimet et al (2002) econometric worklife models with more recent data changes the results in the 2002 paper in any substantive way.

Finding: Overall, the worklife expectancy estimated using more recent data from 2000-2013 is shorter then in the earlier time period (1992-2000) data set. This is true for younger worker (18-early 40’s); younger workers from the more recent cohorts have a shorter expected work life then younger workers in the earlier cohorts.  Conversely, while older workers in their 40s and 50s have a slightly longer worklife expectancy in the later time period data set. We are in the process of determining the statistical significance of these differences.

Notes:

The econometric model described by Millimet  et al (2002) and logistic regression equations by gender and education are used to calculate the worklife expectancy estimates.   The worklife model iin the left panel of the table is estimated using matched CPS cohorts from 1992–2000 time period as described in the Millimet et al. (2002) paper.   The model on the right panel is estimated using data from 2001-2013.

The logistic equation includes independent variable for age, age squared, race, race by age interaction, race by age interaction squared, marital status, martial status by age, occupation dummies, year and year dummies.

The model is first estimated separately for each gender and education level combination for active persons.  The model is then estimated again for inactive persons.  The educational attainment variables used to estimate our model differ from that of Millimet et al. (2002)   In our model, only individuals whose highest level of attainment is high school are included in the high school category.  Millimet et al (2002) includes individuals with some college in the high school category.

Big Data. Bureau of Labor Statistics. Survey data. Employment Big Data.  Those are all things that calculating worklife expectancy for U.S. workers requires.  Worklife expectancy is similar to life expectancy and indicates how long a person can be expected to be active in the workforce over their working life.  The worklife expectancy figure takes into account the anticipated to time out of the market due to unemployment, voluntary leaves, attrition, etc

Overall the goal of our recent work is to update the Millimet et al (2002) worklife expectancy paper and account for more recent CPS data. Their paper uses data from  the 1992 to 2000 time period. Our goal is to update that paper using data from 2000 to 2013. The main goal of the paper is to see if estimating the Millimet et al (2002) econometric worklife models with more recent data changes the results in the 2002 paper in any substantive way

As for the results, overall there are several findings. First we were able to create a match CPS data set of 201,797 individuals where as the Millimet et al. (2002) found 200,916 matched individuals.

Overall we match their results very closely as well.  For example Millimet et al. (2002) found that a male who was 26 years old with a less than a high school education had a 27.27 years WLE remaining while we found that person had 26.319 years remaining based on our replication of their work. They found that the same age person with a high school had 32.89 years remaining while we found 32.728 years remaining. The replication was particularly good for both less than high school and high school levels of educational attainment.

The WLE  numbers are close but not quite as close for college and some college. This is primarily due to the fact that we use different definitions of some college and college then Millimet et al. (2002)  did in their 2002 paper

Notes:

The econometric model described by Millimet  et al (2002) and logistic regression equations by gender and education are used to calculate the worklife expectancy estimates.   The model is estimated using matched CPS cohorts from 1992–2000 time period as described in the Millimet et al. (2002) paper.  The logistic equation includes independent variable for age, age squared, race, race by age interaction, race by age interaction squared, marital status, martial status by age, occupation dummies, year and year dummies.  The model is first estimated separately for each gender and education level combination for active persons.  The model is then estimated again for inactive persons.

Big Data. Bureau of Labor Statistics. Survey data. Employment Big Data.  Those are all things that calculating worklife expectancy for U.S. workers requires.  Worklife expectancy is similar to life expectancy and indicates how long a person can be expected to be active in the workforce over their working life.  The worklife expectancy figure takes into account the anticipated to time out of the market due to unemployment, voluntary leaves, attrition, etc.

Overall the goal of our recent work is to update the Millimet et al (2002) worklife expectancy paper and account for more recent CPS data.

The data for all years is shown below.  Ultimately there were over 590,000 data points used in the analysis.

Notes:

The CPS data was matched using the algorithm similar to Millimet et al (2002) and Peracchi and Welch (1995).  Households in rotation 1-4 were matched using the household identifier number to the same household in rotations 5-8 of the following year. Individuals had to have the same sex, race and be a year older in rotation 5-8 to be determined a match.

Big Data. Bureau of Labor Statistics. Survey data. Employment Big Data.  Those are all things that calculating worklife expectancy for U.S. workers requires.  Worklife expectancy is similar to life expectancy and indicates how long a person can be expected to be active in the workforce over their working life.  The worklife expectancy figure takes into account the anticipated to time out of the market due to unemployment, voluntary leaves, attrition, etc.

Overall the goal of our recent work is to update the Millimet et al (2002) worklife expectancy paper and account for more recent CPS data. Their paper uses data from  the 1992 to 2000 time period. Our goal is to update that paper using data from 2000 to 2013. The main goal of the paper is to see if estimating the Millimet et al (2002) econometric worklife models with more recent data changes the results in the 2002 paper in any substantive way.

Our approach is two fold.  First we matched the BLS data cohorts based on the Millimet et al. (2002) and Peracchi and Welch (1995) papers. In a nutshell the CPS matching routine involves matching incoming and outgoing cohorts across a given year.  Once the data is matched, we then look at the work status of the individuals to determine if they were active or in active across the year that they were interviewed by the BLS. . We were able to create a match CPS data set of 201,797 individuals where as the Millimet et al. (2002) found 200,916 matched individuals.

Notes:

The CPS data was matched using the algorithm similar to Millimet et al (2002) and Peracchi and Welch (1995).  Households in rotation 1-4 were matched using the household identifier number to the same household in rotations 5-8 of the following year. Individuals had to have the same sex, race and be a year older in rotation 5-8 to be determined a match.

In the stats world there is somewhat of a debate going on regarding which statistical analyses programs are “better”.  Of course, the answer always depends on what you use it for.  Some like the open-source, developing nature of R.  While others like the established and tried STATA.

In the world of labor and employment economics and in ligation matters that require data analysis of large sets of data, STATA wins hands down.  However, the open source nature of R is appealing in some settings; but the many decades of pre-written (and de bugged) programs make STATA the best choice in most employment and wage and hour cases that require analysis of large data sets.  Performing basic tabulations and data manipulations in R requires many lines of code while STATA often has the command built in.

Here are some interesting snippets from the web on the R v STATA debate:

http://www.researchgate.net/post/What_is_the_difference_between_SPSS_R_and_STATA_software

The main drawback of R is the learning curve: you need a few weeks just to be able to import data and create a simple plot, and you will not cease learning basic operations (e.g. for plotting) for many years. You will stumble upon weirdest problems all the time because you have missed the comma or because your data frame collapses to a vector if only one row is selected.

However, once you mastered this, you will have the full arsenal of modern cutting-edge statistical techniques at your disposal, along with in-depth manuals, references, specialized packages, graphical interface, a helpful community — and all at no cost. Also, you will be able to do stunning graphics.

 

http://forum.thegradcafe.com/topic/44595-stata-or-r-for-statistics-software/

http://www.econjobrumors.com/topic/r-vs-stata-is-like-a-mercedes-vs-a-bus

The Center for Medical and Medicaid Services (CMS) new Open Payments database shows the consulting fees, research grants, travel and other reimbursements made to medical industry in 2013

There are 2,619,700 payments in the CMS data made to 356,190 physicians.   The average payment made to physicians was $255.22.   The median payment was $15.52

Table 1: Summary of Payments – STATA  output

The physicians received an average total  of $1,877.11 in payments.  The median total payment for the 356,190 physicians in the data was $94.15

Table 2: Summary of Payments – STATA  output

Below is the STATA code for the results:
count gen
bysort phy: gen hj =_n
bysort physician_p: gen hj =_n
sum hj, det
count if hj==1
sum tot, det
bysort physician_p: egen hj2 = tot(total)
bysort physician_p: egen hj2 = total(total_am)
sum hj2
sum hj2, det
sort physcian_p
sort physician_p
sum hj2 if _n==1, det
sum hj, det
sum hj2 if hj==1, det

Many employer time records are still in paper format that is not easily machine readable.  Analyzing these data in a wage and hour case typically requires manual entry of the records.  However, entering in all the time records is not always feasible.  In these situations a statistical random sampling of records can be useful and informative.   A solid statistical sampling alows the researcher to calculate error rates which are useful when making inferences concerning the time records.