Mistake in GarchMidas user's guide. Example is not replicable

Hello,

I ran the code in example 1 to replicate the examples shown in the user guide and I get exactly your results for the RV and forecast exercise. The example can be found on pages 7 (second half) to 8 of the attached GarchMidas user's guide. However, I get completely different results if I run the code regarding the INDPRO dataset. I get following message and results (see also the Figure in the attachment):

Warning: Covariance matrix of estimators cannot be computed precisely due to inversion difficulty. > In GarchMidas (line 311) In my_example (line 86) Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 6.977894e-18. > In GarchMidas (line 313) In my_example (line 86)

I used exactly the same code and data but the example regarding the INDPRO dataset does not work. Hence, I conclude that the code or the assumptions behind it is wrong.

This example is very important to me. Could anybody help me solve this problem and correct the code? Thank you very much.

[JACK Cox]

I am also using this example and user guide as the basis of instruction for a research project and have encountered some issues. Did you manage to correct the code?

[Walter Roberson]

This appears to refer to the File Exchange contribution https://www.mathworks.com/matlabcentral/fileexchange/45150-midas-matlab-toolbox

[MUHAMMAD SHOAIB]

i also same issue......appear the following message.

Warning: Covariance matrix of estimators cannot be computed precisely due to inversion difficulty.

> In GarchMidas (line 311)

In appGARCHMIDAS1 (line 23)

Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND = 1.012192e-17.

> In GarchMidas (line 313)

In appGARCHMIDAS1 (line 23)

KINDLY HELP