## Inverse of a covariance matrix (loop)

on 25 Apr 2013

### Andrei Bobrov (view profile)

Hi all,

I am stuck to create a loop which yields inverse of covariance matrices.

Data description:

I have the returns of three risky assets: mkt, hml and mom, from nov 3, 1926 up to dec 31, 2012.

For each year (so starting from Nov 3, 1927)I want the inverse covariance matrix for the three risky assets.

The dates are described (thanks Andrei) by the following code:

```d = [19261103; 20121231];
ddte = datenum(num2str(d),'yyyymmdd');
ndte = (ddte(1):ddte(2))';
t = weekday(ndte);
ndte = ndte(t ~= 1 & t ~= 7);
yourdata = [date,mkt,hml,mom];
[yy,mm,dd] = datevec(yourdata(:,1));
ymd = [yy,mm,dd];
im = mm == 11 & dd >= 3;
ii = strfind([~im(1),im(:)'],[0 1]);
So I am stuck what do I have to add to retrieve the inverse covariance matrices per year.
```

Hopefully someone can help me out.

Thanks!

I adjusted the code with cov in it, but it does not yield the desired results:

```d = [19261103; 20121231];
ddte = datenum(num2str(d),'yyyymmdd');
ndte = (ddte(1):ddte(2))';
t = weekday(ndte);
ndte = ndte(t ~= 1 & t ~= 7);
yourdata = [date,mkt,hml,mom];
[yy,mm,dd] = datevec(yourdata(:,1));
ymd = [yy,mm,dd];
im = mm == 11 & dd >= 3;
ii = strfind([~im(1),im(:)'],[0 1]);
sb = zeros(numel(ndte),1);
sb(ii) = 1;
sbc = cumsum(sb);
t = sbc > 0 & sbc ~= max(sbc);
sbb = sbc(t);
sb1 = find(sb(t));
wdta = yourdata(t,:);
[r, c] = ndgrid(sbb,1:size(wdta,2)-1);
out1 = accumarray([r(:) c(:)],reshape(wdta(:,2:4),[],1),[],@cov);
out = [ymd(ii(1:end-1),:),out1] ;
What do I do wrong?
```

### Andrei Bobrov (view profile)

on 25 Apr 2013

Try this is code:

```d = [19261103; 20121231];
ddte = datenum(num2str(d),'yyyymmdd');
ndte = (ddte(1):ddte(2))';
t = weekday(ndte);
ndte = ndte(t ~= 1 & t ~= 7);
```
```yourdata = [ndte,mkt,hml,mom];
```
```[yy,mm,dd] = datevec(yourdata(:,1));
ymd = [yy,mm,dd];
im = mm == 11 & dd >= 3;
ii = strfind([~im(1),im(:)'],[0 1]);
sb = zeros(size(yourdata,1),1);
sb(ii) = 1;
sbc = cumsum(sb);
t = sbc > 0 & sbc ~= max(sbc);
sb1 = diff(find([sb(t);1]));
wdta = yourdata(t,:);
s = size(wdta,2) - 1;
ydcell = mat2cell(wdta(:,2:end),sb1,s);
out = cellfun(@(x)cov(x)\eye(s),ydcell,'un',0);
```

### Kevin van Berkel (view profile)

on 25 Apr 2013

Andrei you are a legend. Works perfect, thank you very much!

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi