Re: st: RE: Save results from flexible frequency table in 'matrix like' format

[Nick Cox <njcoxstata@gmail.com>]

There is no Mata in Red Owl's code.

More importantly,

Did you try his code? It's reproducible. There is no need to guess
what it does.

What would a multidimensional frequency table look like? You haven't
explained what you want to produce.

Nick
njcoxstata@gmail.com


On 18 July 2013 10:45, Milena Przheska
<milena.przheska@cosmicdevelopment.com> wrote:
> Hi Red,
>
> Thank you very much for this algorithm.
>
> I am quite new to Stata and not yet very familiar with Mata, so I was
> wondering how should your code be extended to include more than one
> variable in order to create multidimensional frequency table, because
> the way I understand it, it creates vector, not a matrix.
>
> Once again, thank you for your response, I appreciate it.
>
> Best,
> Milena Przheska
>
> On Wed, Jul 17, 2013 at 8:56 PM, Red Owl <rh.redowl@liu.edu> wrote:
>> The example below may not be the most efficient way to create a vector of frequencies, but I believe it works.
>>
>> sysuse auto, clear
>> levelsof mpg, local(mpglevels)
>> local lvlsnum = wordcount("`mpglevels'")
>> matrix FREQMAT = J(`lvlsnum',1,.)
>> local lvlscount = 0
>> foreach lvl of local mpglevels {
>>    local lvlscount = `lvlscount' + 1
>>    quietly tabstat mpg if mpg == `lvl', stat(count) save
>>    mat A = r(StatTotal)
>>    mat FREQMAT[`lvlscount',1] = A[1,1]
>>    }
>> mat colnames FREQMAT = Frequency
>> mat rownames FREQMAT = `mpglevels'
>> matlist FREQMAT
>>
>> Red Owl
>> (Yes, that's my real name)
>>
>>
>>>> Dear members of Statalist,
>>>>
>>>> I have not been able to find appropriate way to save and export
>>>> results from flexible frequency tables, so I kindly ask for your help.
>>>>
>>>> I would like to save for later use, a frequency table with multiple
>>>> categorical variables, just like it appears in the result - like a
>>>> matrix. Also, I need to keep the names of the variables to mark rows
>>>> and columns.
>>>>
>>>> The dataset consists of many observations and mainly categorical
>>>> variables (with some of them having tens of categories).
>>>>
>>>> I have tried the following ways to save the result after creating
>>>> flexible frequency table:
>>>>
>>>> Copying and pasting the table into word is not good because the
>>>> resulting table is very large;
>>>>
>>>> Replacing the original data set with the frequencies does not give me
>>>> the matrix like format of the table;
>>>>
>>>> Collapsing the data after frequency also does not work well, because
>>>> some of the variables are categorical and need to remain so, while
>>>> frequencies are summed;
>>>>
>>>> The -matcell- option only worked with two-way frequency tables;
>>>>
>>>> -return- and -ereturn- did not contain any matrices.
>>>>
>>>> I would very much appreciate your suggestions and I thank you in advance.
>>>>
>>>> Regards,
>>>> Milena Przheska
>> *
>> *   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/