hanssmail / quantq Goto Github PK

Excuse me. In the directory "data", I can't find the dataset that used in the book.

Hi,
there's a small error in the output of t5,t6 on p.123: The output displayed shows t1,t6 rather than t5,t6

and there's an error in either the explanation of the example for or the window used in 7.2.10 for the window join. p.140

Currently reading/working through the book and I really enjoy it.

Thanks

Regarding .quantQ.ols.olsTab
under tabStats
"f"$count[y] should read "f"$count[y] because y is referencing a variable under ols.fit, which counts all nonmissing observations in the table.

It seems that quantQ_nn.q does not contain the definitions of .quantQ.nn.funcNN and .quantQ.nn.funcErrNN, therefore when .quantQ.nn.modelNN is called it yields an evaluation error. Adding this in the library fixes the problem and the function runs, reproducing the example in page 451 of the book

When using this function I get an evaluation error at "log[2*acos-1]". The function seems to be missing a separation between "acos" and "-1", as it is actually printed in page 250 of the book.
Thanks a lot for the great library and book

In page 254 of the book it states that olsTab is nested under namespace "ols" but in the current version of the library seems to be under quantQ directly (quantQ.olsTab instead of quantQ.ols.olsTab as in the book)
Best regards

Function returns 2 normal variates. In Chapter 14 the results are sampled as e.g. 100?.quantQ.simul.genBoxMuller[] (sampling 100 values from a list of two normal variates). I suspect the intention is that .quantQ.simul.genBoxMuller[n] return n random variates or similar

I think the formula is EMA_i(λ) = x_i ✕ (1-λ) + EMA_i-1 ✕ λ
Not x_i-1 in the first term.

Thanks,

Both in the book and in the library the ols.RMSE function is implemented using the wavg function, but this seems to me incorrect, since wavg is according to the documentation (sum x*y) % sum x, so it is dividing by sum t instead of count t. This actually makes possible to have negative values depending on the order of model 0 and 1. In my opinion a simple solution could be to use sqrt avg t * t: tabModel...
Best regards

The function defined in page 487 .quantQ.trees.runRule is missing from the library file, making predictOnTree fail. By adding it as defined in the book the example from the book is reproduced.

Hi, in page 255 function ols.fStatistics has a count[] function that is not necessary and is actually corrected in the library. I mention this in case you want to add this to the documentation of typos from the book
Best regards

I had to do a few adaptations of the code in order to reproduce the results of example 14.3.1.3 on VAR models. In general all had to do with overloaded functions like ts.Y, ts.zt and ts.Z.

• For ts.Y there are two versions of the same function, the only difference being the evaluation of y[;0] or y[0;]. The problem seems to be that VAR functions like ts.varOrder call the function with the wrong index order.

• For ts.zt and ts.Z the overloaded versions have an extra parameter "flag", but functions like ts.varOrder call the curried version which return a function and therefore the function fails.

In all these cases my solution has been to rename the overloaded functions, I don't know if there is a way to impose the use of one of the specific versions, or there should be a special namespace for the VAR utility functions.
Best regards