Hi Jan,

I noticed a small bug in xtset2.ado, line 90: svmat creates the variable but appends an extra 1 at the end, i.e. Tmat1 instead of Tmat. This makes xtset2 fail when using set varabbrev off. This behavior also occurs in line 107 with sum.

Best,

Chris

Instead of loop in loop, multiply e (residuals, NxT) with ej where ej is e with jth element to front, i.e. ej = (ej, ej+1,...,eN,e1,...,ej-1). loop over ej and multiply with e. gives N cross correlations, i.e. one row for rho matrix.

advantage: only one loop necessary. faster?
disadvantage: if large T and large N, speed might be slower.
solutions: do test runs and assign threshold when multiplication gets slower.

I keep on getting an error message when I try to run the codes in the help file in Stata. Can you kindly check if there are any bugs (with stata 17)?

Add a check that half of a panel is large enough to run with jackknife

• gap after "Short Run Est."
• Headline misses Mean Group Estimator, says only CS-ARDL in brackets.

only last CD value is saved in r(CD)

min and max of T in unbalanced panels

First thank you for the codes. I checked and compared the outputs of RSS (residual sum of squares), MSS (model sum of squares) and R-squared. There may be something wrong. The formula seems to be R-sq = rss / (rss+mss) . I think either the R-sq formula is wrong, or the names assigned to rss and mss have been mixed up. Thank you very much!

The Dropbox link to an example dataset in the readme file is Broken.

check which selectindex to use. if version < 14 { selectindex() } to auxilary script and replace xtdcce2_selectindex ... with selectindex

Problem with partialling out a constant, try to use svsolve in mata instead of qr solve or force solver to be always the same.

Underlying problem is that Stata matrix inverter might produce different results.

always close to zero because lower part (Yi - Ybar) always small

Problem xtdcce_m_lrcalc assigns varnames for variables as varname_1...rows(beta vector); need to add id variable argument to lr program.

Coefficients for omitted variables are included in mean group regression. User needs to be warned and options should be given to exclude those zeros.

Work required in mean group program.

Shouldn't this be xtdcce2_mm_which2 ? mm_which2 is undefined in my version of moremata.

If option jackknife is used, then the cross sectional averages need to be partial out over the first and the second half of the dataset, rather than over the entire.

See attached code:

clear

set obs 100

gen t = _n

gen x = rnormal()
gen cons = 1
gen y = 2cons + 0.5x + rnormal()

reg y cons , noconst
predict ytilde , res

reg x cons, noconst
predict xtilde, res

reg y x
scalar b1mg = _b[x]
reg ytilde xtilde, noconst
scalar b1mgtilde = _b[xtilde]

** Jackknife

reg y x if t < 50
scalar b1 = _b[x]

reg y x if t >= 50
scalar b2 = _b[x]

reg ytilde xtilde if t < 50, noconst
scalar b1tilde = _b[xtilde]

reg ytilde xtilde if t >= 50, noconst
scalar b2tilde = _b[xtilde]

** Correct partial out
reg y cons if t < 50, noconst
predict y1tilde, res
reg y cons if t >= 50, noconst
predict y2tilde, res

reg x cons if t < 50, noconst
predict x1tilde, res
reg x cons if t >= 50, noconst
predict x2tilde, res

reg y1tilde x1tilde if t < 50, noconst
scalar b12tilde = _b[x1tilde]
reg y2tilde x2tilde if t >= 50, noconst
scalar b22tilde = _b[x2tilde]

display "normal regression"
display 2b1mg - 0.5 * (b1 + b2 )
display "wrong model"
display 2b1mgtilde - 0.5 * (b1tilde + b2tilde)
display "correct model with partial out for each half"
display 2*b1mgtilde - 0.5 * (b12tilde + b22tilde)

estat quits with error message.
Problem is that local with variables for graph is empty because variables are missing in lists to create graphs.
Need fix in xtdcce2 main program and returned variables.