Category Archives: Blog

  • 1

Getting p-values and t-values with asreg

Category:Blog,Stata Programs

Xi asked the following quesiton:

How can I get p-values and t-values using asreg program?

Introduction to asreg

asreg is a Stata program, written by Dr. Attaullah Shah. The program is available for free and can be downloaded from SSC by typing the following on the Stata command window:

ssc install asreg

asreg was primarily written for rolling/moving / sliding window regressions. However, with the passage of time, several useful ideas were conceived by its creator and users. Therefore, more features were added to this program. 

Getting t-values after asreg

Consider the following example where we use the grunfeld dataset from the Stata web server. The dataset has 20 companies and 20 years of data for each company. We shall estimate the following regression model where the dependent variable is invest and independent variables are mvalue and kstock. Let’s estimate the regression model separately for each company, denoted by i .

In the following lines of code, the letters se after comma causes asreg to report the standard errors for each regression coefficient.

webuse grunfeld, clear
bys company: asreg invest mvalue kstock, se

asreg generates the regression coefficients, r-squared, adjusted r-squared, number of observations (_Nobs) and standard errors for each coefficient. This is enough information for producing t-values and p-values for the regression coefficients. The t-values can be generated by:

gen t_values_Cons = _b_cons / _se_cons
gen t_values_mvalue = _b_mvalue / _se_mvalue
gen t_values_kstock = _b_kstock / _se_kstock

Getting p-values after asreg

Getting p-values is just one step away. We need one additional bit of information from the regression estimates, that is the degrees of freedom. This is usually equal to the number of observations minus the number of parameters being estimated. Since we have two independent variables and one constant, the number of parameters being estimated are 3. asreg returns the number of observation in the variable _Nobs. Therefore, the term _Nobs – 3 in the following lines of code is a way to get the degrees of freedom.

gen p_values_Cons = (2 * ttail(_Nobs-3), abs( _b_cons / _se_cons ))
gen p_values_mvalue = (2 * ttail(_Nobs-3), abs( _b_mvalue / _se_mvalue ))
gen p_values_kstock = (2 * ttail(_Nobs-3), abs( _b_kstock / _se_kstock ))

Verify the Results

Let’s estimate a regression for the first company and compare our estimates with those produced by the Stata regress command.

 reg invest mvalue kstock if company == 1
list _b_cons _se_cons p_values_Cons in 1

Full Code

 webuse grunfeld, clear
bys company: asreg invest mvalue kstock, se
gen t_values_Cons = _b_cons / _se_cons
gen t_values_mvalue = _b_mvalue / _se_mvalue
gen t_values_kstock = _b_kstock / _se_kstock
gen p_values_Cons = (2 * ttail(_Nobs-3), abs( _b_cons / _se_cons ))
gen p_values_mvalue = (2 * ttail(_Nobs-3), abs( _b_mvalue / _se_mvalue ))
gen p_values_kstock = (2 * ttail(_Nobs-3), abs( _b_kstock / _se_kstock ))
reg invest mvalue kstock if company == 1
list _b_cons _se_cons t_values_Cons p_values_Cons in 1

  • 4

Ordering variables in a nested regression table of asdoc in Stata

Category:asdoc,Blog Tags : 

In this blog entry, I shall highlight one important, yet less known, feature of the option keep() in nested regression tables of asdoc. If you have not used asdoc previously, this half-page introduction will put on fast track. And for a quick start of regression tables with asdoc, you can also watch this YouTube video.


Option keep()

There are almost a dozen options in controlling the output of a regression table in asdoc. One of them is the option keep(list of variable names). This option is primarily used for reporting coefficient of the desired variables. However, this option can also be used for changing the order of the variables in the output table. I explore these with relevant examples below.


1. Changing the order of variables

Suppose we want to report our regression variables in a specific order, we shall use option keep() and list the variable names in the desired order inside the brackets of option keep(). It is important to note that we have to list all variables which we want to report as omitting any variable from the list will cause asdoc to omit that variable from the output table.


An example

Let us use the auto dataset from the system folder and estimate two regressions. As with any other Stata command, we need to add asdoc to the beginning of the command line. We shall nest these regressions in one table, hence we need to use the option nest. Also, we shall use option replace in the first regression to replace any existing output file in the current directory. Let’s say we want to variables to appear in this order in the output file _cons trunk weight turn. Therefore, the variables are listed in this order inside the keep() option. The code and output file are shown below.

sysuse auto, clear
asdoc reg mpg turn, nest replace
asdoc reg mpg turn weight trunk, nest keep(_cons trunk weight turn)



2. Reporting only needed variables

Option keep is also used for reporting only needed variables, for example, we might not be interested in reporting coefficients of year or industry dummies. In such cases, we shall list the desired variable names inside the brackets of the keep() option. In the above example, if we wish to report only _cons trunk weight , we would just skip the variable turn from the keep option. Again, the variables will be listed in the order in which they are listed inside the keep option.  

sysuse auto, clear
asdoc reg mpg turn, nest replace
asdoc reg mpg turn weight trunk, nest keep(_cons trunk weight)



Off course, we could also have used option drop(turn) instead of option keep(_cons trunk weight) for dropping variable turn from the output table.



  • 0

Exporting ttest results from Stata to Word using asdoc

Category:asdoc,Blog,Stata Programs,Uncategorized


asdoc installation

If you have not already studied the features of asdoc, you can visit this page that lists the table of contents of what asdoc can do. You can also read this one paragraph introduction to asdoc. The following line of code will install asdoc from SSC

ssc install asdoc
help asdoc


Reporting t-tests with asdoc

Before we make the t-test results table for our example data, let us breifly explore the options available in asdoc for making a t-test results table.

Whether it is one-sample t-test or two-sample or other forms, asdoc manages to report the results line by line for each test. asdoc also allows accumulating results from different runs of t-tests. For this purpose, the option rowappend of asdoc really comes handy. With the sub-command ttest , we can use the following options of asdoc to control asdoc behavior.

(1) replace / append

(2) save(filename)

(3) title(text)

(4) fs(#)

(5) hide.

(6) stats()

(7) rowappend.

These options are discussed in detail in Section 1 of asdoc help file. Option stats and rowappend are discussed below:


Option stat()

Without stat() option, asdoc reports the number of observations (obs), mean, standard error, t-value, and p-value with t-tests. However, we can select all or few statistics using the stat option. The following table lists the keywords and their details for reporting the desired statistics.

keyword details
n Number of observations
mean Arithmetic mean
se Standard error
df degrees of freedom
obs Number of observations
t t-value
p p-value
sd standard deviation
dif difference in means if two-sample t-test


Option rowappned

ttest tables can be constructed in steps by adding results of different t-tests to an existing table one by one using option rowappend. There is only one limitation that the t-tests are performed and asdoc command applied without writing any other results to the file in-between.


An example

Suppose we have the following data set with variables r0, r1, r2, r3, and y. The data can be downloaded into Stata by

use, clear

The variables ro-r3 are the numeric variables for which we would like to conduct one-sample ttest whereas variable y is a numeric date variable that tracks years. We wish to conduct a ttest for each of the r0-r3 variables and in each year and make one table from all such tests.


Without using a loop

 asdoc ttest R0==0 if Y==2009, replace title(One Sample t-test Results)
asdoc ttest R1==0 if Y==2009, rowappend
asdoc ttest R2==0 if Y==2009, rowappend
asdoc ttest R3==0 if Y==2009, rowappend
asdoc ttest R0==0 if Y==2010, rowappend
asdoc ttest R1==0 if Y==2010, rowappend
asdoc ttest R2==0 if Y==2010, rowappend
asdoc ttest R3==0 if Y==2010, rowappend
asdoc ttest R0==0 if Y==2011, rowappend
asdoc ttest R1==0 if Y==2011, rowappend
asdoc ttest R2==0 if Y==2011, rowappend
asdoc ttest R3==0 if Y==2011, rowappend
asdoc ttest R0==0 if Y==2012, rowappend
asdoc ttest R1==0 if Y==2012, rowappend
asdoc ttest R2==0 if Y==2012, rowappend
asdoc ttest R3==0 if Y==2012, rowappend

And appreciate the results



1.In the first line of code, we wrote asdoc ttest in the beggining of the line. This is how we use asdoc with Stata commands. We just add asdoc to the beggining of any Stata command and that’s all.

2. We used two options of asdoc in the first line of code: the replace and title(). Replace will replace any existing file with the name Myfile.doc and title will add the specific test as a title to the output file.

3. In the second line of code, we added option rowappend() that will append the results to the existing table in the file Myfile.doc

4. And the process continues untill all ttests are estimated.


  • 7

asdoc: Cutomizing the regression output | MS Word from Stata | Confidence Interval, adding stars, etc.

Category:asdoc,Blog,Uncategorized Tags : 


Version 2.3 of asdoc adds the following features for reporting detailed regression tables.

1. Reporting confidence interval

2. Suppressing confidence intervals

3. Suppressing the stars which are used to show significance level

4. Customization of significance level for stars

These features are discussed in details below. If you have not already studied the features of asdoc, you can visit this page that lists the table of contents of what asdoc can do. You can also read this one paragraph introduction to asdoc. The following line of code will install this beta version of asdoc from our website

net install asdoc, from( replace
help asdoc


Details of the new features

The new features related to creating detailed regression tables with asdoc are discussed below with details. 


1. Confidence interval

I received several emails and comments on blog posts suggesting the addition of confidence intervals (CI) to the detailed regression tables created by asdoc. In version 2.3 onwards, confidence intervals are shown by default. This means that we do not have to add an additional option to report CI. See the following example. 

sysuse auto, clear
asdoc reg price mpg rep78 headroom trunk weight length turn , replace



2. Suppressing the confidence interval

If confidence intervals are not needed, we can use option noci. For example

asdoc reg price mpg rep78 headroom trunk weight length turn , replace noci


3. Suppressing stars

Similarly, if we are not interested in reporting significance stars, we can use option nostars. For example, 


4. Setting custom significance level

The default significance levels for reporting stars are set at : *** for p-values <=0.01; ** for p-values <=0 .05, and * for p-values <=0.1. However, now we can set our own levels for statistical significance using option setstars. An example of setstars option looks like:

setstars(***@.01, **@.05, *@.1)

As we can see from the above line, setstars separates each argument by a comma. Each argument has three components. The first component is the symbol (in our case it is *) which will be reported for the given significance elve. The second component is the @ sign that connects the significance level with the symbol. And the third component is the value at which the significance level is set. So if we want to report stars such that

* for p-value .001 or less
** for p-value .01 or less
*** for p-value .05 or less

We shall write the option setstars as

setstars(*@.001, **@.01, ***@.05)

Continuing with our example, let us use the above option to report our defined level of stars.

asdoc reg price mpg rep78 headroom trunk weight length turn , replace setstars(*@.001, **@.01, ***@.05)



  • 6

asdoc: Exporting customized descriptive statistics from Stata to MS Word / RTF


Osama Mahmood has asked : 

If I want to report 25th and 75th percentiles for variables through asdoc, then how would I do that? And what if I do not want to report the Min and Max?

Answer: In this YouTube video, I have shown various methods in which descriptive statistics can be reported using asdoc. What Osama has asked for is possible with the customized descriptive statistics using the stat() option of asdoc. Using option stat(), we can choose from the following statistics. Each of the bold words in the following list represents the control word that can be used to report the required statistic.

N Number of observations
mean Arithmetic mean
sd Standard deviation
semean Stanard error of the mean
sum Sum / total
range Range
min The smallest value
max The largest value
count Counts the number of non-missing observations
var Variance
cv Coefficient of variation
skewness Skewness
kurtosis Kurtosis
iqr Interquartile range
p1 1st percentile
p5 5th percentile
p10 10th percentile
p25 25th percentile
p50 Median or the 50 percentile
p75 75th percentile
p99 99th percentile
tstat t-statistics that the given variable == 0


Example 1: Mean, sd, 25th percentile, median, and 75th percentiles

 sysuse auto
asdoc sum, stat(mean sd p25 p50 p75) replace


Example 2: Mean, sd, 25th percentile, median, and 75th percentiles, range, t-statistics

 asdoc sum, stat(mean sd p25 p50 p75 range tstat) replace

  • 2

asdoc: Export matrix to MS Word | the case of xttab command in Stata


asdoc provides a variety of ways in which results from various Stata commands can be exported to MS Word or an RTF file. In this blog post,  I show how to export a Stata matrix to MS word. Usually, Stata commands leave results in r() or e() macros and sometimes in a Stata matrix. Consider the example of xttab command.  xttab is a generalization of tabulate oneway. It performs one-way tabulations and decomposes counts into between and within components in panel data. The command returns results in the r(results) matrix which we can then send to MS word.


The syntax

asdoc follows the following syntax for exporting matrix to a word document.

asdoc wmat, matrix(matrix_name) [rnames(row names) cnames(row names) replace append other_options]



wmat is the command name – an abbreviation for writing matrix. Option matrix() is a required option to get the name of an existing matrix. Option rnames() and cnames() are optional options to specify row names and column names of the matrix. If these options are left blank, existing row and column names of the matrix are used. Other options of asdoc can also be used with wmat. For example, replace will replace an existing output file, while append will append to the existing file. fs() sets the font size, while option title() can be used to specify the title of the matrix in the output file.

An example: The case of xttab command

The dataset that we shall use is from the help file of xttab.

webuse nlswork
xtset id year
xttab race
mat T = r(results)
asdoc wmat, mat(T) replace



1. The first line downloads the example data

2. The second line declares the data as panel data

3. The third line tabulates the race variable

4. The fourth line creates a matrix with the name T from the xttab command

5. The fifth line writes the T matrix to a Word file. wmat is a sub-command in asdoc for writing matrix data to the output file. The two words after command are options of asdoc. The first option tells asdoc about the name of the matrix that has to be exported. The second option tells asdoc to replace any existing output file.

asdoc produces the following Table.

Results Table
































Over a grouping variable?

If we wished to do the above for each category of the grouping variable msp, that has two categories i.e., 0 and 1, we can use the if qualifier and append the results to the same file. So

xttab race if msp == 1
mat T = r(results)
asdoc wmat, mat(T) replace title(When msp == 1)
xttab race if msp == 0
mat T = r(results)
asdoc wmat, mat(T) title(When msp == 0)


When msp == 1





































When msp == 0





































  • 13

asdoc: using option row for creating customized tables row by row in Stata | MS Word

Category:asdoc,Blog Tags : 


Option row is a new feature in version 2.0 of asdoc. This feature allows building tables in pieces. That is good news for those who want to make highly customized tables from Stata output.

This feature can be considered an advanced topic and might not be good for Stata beginners. With many other Stata commands, using asdoc is exceptionally easy. You can read this concise blog post for some basic examples of using asdoc. However, if you are already familiar with Stata macros and results returned in r() and e() macros, then you should continue reading this post.

How does option row work?

Option row allows building a table row by row from text and statistics. In each run of asdoc with option row, a row is added to the output table. The syntax for using this option is given below:

asdoc, row(data1, data2, data3, ...)

As shown above, we shall type nothing after the word asdoc. Therefore, all other arguments of the command come after the comma. The first required option is row(data1, data2, …). Here data1, data2, … can be either a numeric value, string, or both. Within the brackets after option row, each piece of data should be separated by the character comma and hence it will be written to a separate cell in the output table. If a cell is empty, then each comma should be accompanied by a backslash that is  “,\”


We can use the following options when using option row. dec(): for specifying the number of decimal points. If not used, the default is to use three decimal points. An example of using this option could be dec(2) for using two decimal points. title() : This will add a title to the table. This option works only when the row option is used for the first time in the creation of a table. For example, title(Descriptive Statistics). replace: this option will replace any existing file. Without option replace, the default is to append results.

save(): This will save file with the specified name. For example, save(Table 1) will save the file with the name Table 1. 

A simple example

To understand how does the option row work, let us write first the table column title and then some data. Let us create a table that has four columns. The columns are named as KP, Sindh, Baluchistan, and Punjab. We shall write the table title as Provincial GDP of Pakistan.  So the first row is the header row

 asdoc, row(Years, KP, Sindh, Baluchistan, Punjab) title(Provincial GDP of Pakistan over years) replace

The above line of code generates the table title and the header row. Please note that we also included Years in the table columns because we shall report the provincial GDP over years, therefore we need one additional column for displaying the year labels in the first column. Now let us continue writing to this table. Make sure that you close the Word file before writing additional rows to it.

asdoc, row(1999, 2500.55, 4000.35, 1000.21,  5500.74) dec(2)

In the second line of code, we did not write replace as we wanted to append the results to the same file “MyFile.doc” and we also skipped the title option. We used option dec(2) to report two decimal points with numeric values. We can continue writing additional rows to this table.

asdoc, row(200, 2600.25, 4500.35, 1100, 5700.87) dec(2)

Collecting stats with option accum

We can create a table from text and statistics that are collected from different Stata commands. There is one challenge to developing such a flexible table with option row – that a given row has to be written in one go. So once a row is written, no further cells can be appended to the same row. This means that we need to first collect all the required bits of information before writing a row. Collecting and holding these bits of information can be tricky or too time-consuming. To facilitate this process, asdoc offers option accum(data1 data2,…). The word accum is an abbreviation that I use for accumulate. The syntax of this option is given below:

asdoc, accum(data1 data2 data3 data4 data5 ...) [ dec(#) show ]

Actually, the above command can be run as long as the limit of the global macro to hold data is not reached. The above command will accumulate text and statistics from different runs of asdoc and hold them in the global macro ${accum}. Once we have accumulated all the needed bits of information in the global macro, then its contents can be written to the Word table with option row. Option show can be used to show the contents of the global macro ${accum}. Assume that we want to build an odd table that presents the number of observations, mean, and standard deviation for two variables in two different time periods. The researcher wants to follow the following format:

webuse grunfeld, clear

asdoc, row( \i, \i, invest, \i, \i, kstock,\i) replace

asdoc, row( Periods, N, Mean, SD, N, Mean, SD)

sum invest if inrange(year , 1935, 1945)

asdoc, accum(`r(N)', `r(mean)', `r(sd)')

sum kstock if inrange(year , 1935, 1945)

asdoc, accum(`r(N)', `r(mean)', `r(sd)')

asdoc, row( 1935-1945, $accum)


1. The second row of our required table reveals that a total of 7 cells are needed, this is why we created 7 cells in the first line of code. The text ” \i,” is a way of entering an empty cell. We entered empty cells so that the variables names invest and kstocks are written in the middle of the table.

2. The second line of code writes the table header row.

3. The third line finds summary statistics. We shall collect our required statistics from the macros that are left behind in r() by the sum command.

4. The fourth line accumulates the required statistics for our first variable invest

5. We are not yet writing the accumulated statistics to the Word file. So we find statistics for our second variable kstocks in the fifth line.

6. We again accumulate the needed statistics for our second variable in the sixth line.

7. Since our row of required statistics is now complete, we write the accumulated statistics and the first-row label, i.e, 1935-1945 to our Word file. Let us write one more row to the table. This time, the statistics are based on years 1946-1954

sum invest if inrange(year , 1946, 1954)

asdoc, accum(`r(N)', `r(mean)', `r(sd)')

sum kstock if inrange(year , 1946, 1954)

asdoc, accum(`r(N)', `r(mean)', `r(sd)')

asdoc, row( 1946-1954, $accum)

  • 11

Research Topics in Finance: Asset Pricing

Category:Blog Tags : 


Investor sentiment: Does it augment the performance of asset pricing models?

Mispricing and the five-factor model

Size, value, profitability, and investment: Evidence from emerging markets

4 Noisy prices and the Fama–French five-factor asset pricing model

5 Cross-sectional tests of the CAPM and Fama–French three-factor model

6 Decomposing the size, value and momentum premia of the Fama–French–Carhart four-factor model

7 Monday effect in Fama–French’s RMW factor

8 Digesting anomalies in emerging markets: A comparison of factor pricing models

9 Q-theory, mispricing, and profitability premium

10 Limits of arbitrage and idiosyncratic volatility

11 Is size dead? A review of the size effect in equity returns

12 Market states and the risk-based explanation of the size premium

13 Market volatility and momentum

14 A risk-return explanation of the momentum-reversal “anomaly”

15  Time-varying risk, mispricing attributes, and the accrual premium

16 Bayesian tests of global factor models

17 Model comparison tests of linear factor models in stock returns

18 Multi-factor asset pricing models: Factor construction choices and the revisit of pricing factors

19 Idiosyncratic volatility in the Asian equity market

20 What global economic factors drive emerging Asian stock market returns?

  • 2

asdoc : Easily create Summary Stats in Stata and send it to MS Word: A video example

Category:Blog Tags : 

In this video post, I show the use of asdoc for the different type of summary statistics in Stata and sending them to MS word. Examples given in this video include:

  1. Default summary statistics that include the number of observations, mean, standard deviation, minimum, and maximum
  2. Detailed summary statistics that include no. of observations, mean, standard deviation, 1st percentile, median, 99th percentile, skewness, and kurtosis
  3. Customized summary statistics
  4. Controlling the number of decimal points
  5. Creating new files
  6. appending to existing files

I would really appreciate if you comment on this video on Youtube and subscribe to my channel if you have not already done that.

  • 4

Dropping i.dummies from regression | asdoc | Word | Stata

Category:asdoc,Blog,Stata Programs Tags : 

Questions: I have time and location dummies which I want to include in the regression, but do not want to report them in the regression nested tables created with asdoc. How can I do that?

If you have not already installed asdoc, you can install it from SSC by typing the following in the Stata command window:

ssc install asdoc

Let’s use an example data set.

use, clear

This dataset has four main independent variables, named as x1, x2, x3, x4 and a set of possible dummy variables that will be constructed from the variable year (from 2001-2005) and location (from 1-3).  Let us estimated the following regression:

asdoc reg y x1 x2 x3 i.year i.location, nest drop(i.year i.location) replace

asdoc reg y x1 x2 x4 i.year i.location, nest drop(i.year i.location)


In the above two lines, we have estimated two regressions and sent their output to a Word file.  In the first line, we estimated a regression with the three main independent variables x1, x2, and x3 and included the year and location dummies on the fly. The option nest will create a nested regression table. The option drop(i.year i.location) drops these dummy variables from the regression table, however, they are included in the main regression. The two lines produce the following regression table in MS Word. 

[mc4wp_form id=”1409″]