Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down at the end of May, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Arthur Boman <boman@berkeley.edu> |

To |
<statalist@hsphsun2.harvard.edu> |

Subject |
Re: st: how do we jointly test coefficients from different regressions? |

Date |
Tue, 19 Mar 2013 12:07:53 -0700 |

Jorge, When I say 'allow the constant' I just meant that the regression has a constant in it and is not forced to pass through the origin. But now that you mention it... Yes I need to allow the constant to vary over different y variables. I will have to do a lot of reading and study to know what this means. If I trim your code to exclude generation of the dataset, I guess I get the following. I have also tried to include the facts that 1. there is more than just x2, there is also x3 and x4 and that 2. there are y1... y25: clear reshape long y, i(year) j(var) * Regress the stacked y on x, different coefficients by var. * Allowing the constant to change over different y variables: reg y c.(x1 x2 x3 x4)##i.var * Test coefficients on x1. Notice this can be run for any number of variables. qui test 1.var#c.x1=0 forv i=2(1)$nvar { qui test `i'.var#c.x1=1.var#c.x1, accum } test Did I erase anything that I should not have? On Tue, 19 Mar 2013 00:38:36 -0400, Jorge Eduardo Pérez Pérez <perez.jorge@ur.edu.co> wrote: > This is an example of the "stacking" approach using reshape. I don't > understand what you mean when you say you want to allow the constant, > but I contemplate it the example below. > > * ----------------- Begin code --------------------------------------- > clear > version 12 > * This just generates some random data. You do not need to do this. > set obs 68 > set seed 135 > * x variables > gen x1=rnormal() > gen x2=rnormal() > * time > gen time=_n > * 3 y variables (this could be any number of variables, i.e. 25) > glo nvar=3 > * Coefficients. I am setting 0 for x1 and 1 for x2 > forv i=1(1)$nvar { > glo b`i'1=0 > glo b`i'2=1 > } > forv i=1(1)$nvar { > gen y`i'=${b`i'1}*x1+${b`i'2}*x2 > } > * Now I have a data set like yours. > * Reshape, same as stacking y into 1 vector > reshape long y, i(time) j(var) > * Add an error term for the regression. You do not need to do this. > gen e=0.1*rnormal() > replace y=y+e > * Regress the stacked y on x, different coefficients by var. > * Constant is not varying over the different y variables > * If you want to allow the constant to vary, use reg y c.(x1 x2)##i.var > reg y c.(x1 x2)#i.var > * Test coefficients on x1. Notice this can be run for any number of > variables. > qui test 1.var#c.x1=0 > forv i=2(1)$nvar { > qui test `i'.var#c.x1=1.var#c.x1, accum > } > test > * --------------------- End code ---------------------------------------- > > Hope this helps, > > Jorge Pérez. > _______________________ > Jorge Eduardo Pérez Pérez > > > On Mon, Mar 18, 2013 at 11:55 PM, Arthur Boman <boman@berkeley.edu> wrote: >> Hello, >> >> I am working on a joint test. The test is NOT of the standard f-test >> form: >> >> y = a*x1 + b*x2+ c*x3 +d*x4, and then testing the null whether a=b=c=0. >> >> The test is of the form: >> >> y1= a*x1 + f*x2 and y2= b*x1 + g*x2 and y3= c*x1 + h*x2 and testing >> the >> null whether a=b=c=0 >> >> I want to allow the constant. >> >> I have looked a lot and cannot figure out how to do in Stata. >> >> y1, y2, y3, x1, x2 are time series data by year... one value per year. I >> have data for all five of those variables for each of 68 consecutive >> years. >> I don't have data for any of them for any other years. >> >> Someone suggested I stack (y1, y2, y3) into a column vector. I dont get >> how that would work and cannot ask the person. >> >> Thanks, >> Arthur >> >> (More background: y1, y2, and y3 are portfolio returns by year. I want >> to >> test the hypothesis that x1 is not a priced factor in ANY of the >> portfolios >> (i.e. that the coefficient on x1 is zero for ALL portfolios). x2 is just >> another factor in my asset-pricing model. There are actually 25 >> portfolios, not just three. I will be testing whether we can reject the >> null hypothesis that all of the 25 coefficients are zero.) >> >> >> >> >> >> >> >> >> * >> * For searches and help try: >> * http://www.stata.com/help.cgi?search >> * http://www.stata.com/support/faqs/resources/statalist-faq/ >> * http://www.ats.ucla.edu/stat/stata/ >> >> > > > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/faqs/resources/statalist-faq/ > * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/

**References**:**Re: st: how do we jointly test coefficients from different regressions?***From:*Jorge Eduardo Pérez Pérez <perez.jorge@ur.edu.co>

- Prev by Date:
**st: QMLE** - Next by Date:
**Re: st: QMLE** - Previous by thread:
**Re: st: how do we jointly test coefficients from different regressions?** - Next by thread:
**st: Stata command to compare linear combinations of predicted values after regression** - Index(es):