Spread monthly data into days

Hello,

I am new to Matlab and I am trying to spread monthly data into a table that is indexed daily. For example:

Date Data

01/1982 0.15

02/1982 -0.02

03/1982 -0.05

What I would like to achieve...

Date Data

1/1/1982 0.15

1/2/1982 0.15

1/3/1982 0.15

1/4/1982 0.15

....

1/31/1982 0.15

2/1/1982 -0.02

2/2/1982 -0.02

2/3/1982 -0.02

2/4/1982 -0.02

I other words, I am trying to convert a table with 240 rows (20 years in months) to one with 7300 rows (20 years in days).

Is this easily doable in matlab? Greatly appreciate any advice.

Answer 1

[Kelly Kearney]

One way to do the expansion... convert your monthly dates to the first of each month, then use dateshift to check which month each of your days match:

% Some placeholder data... you'll have to read your tables as appropriate
tmon = datetime(1982, 1:3, 1);
ymon = [0.15 -0.02 -0.05];
tday = datetime(1982,1,1) + days(0:89);

% Match up days to months
[~,loc] = ismember(dateshift(tday, 'start', 'month'), tmon);
yday = ymon(loc);

[Ajay Gupta]

Thank you! It worked but now I have to complicate the problem.

The dateshift method you showed worked for one column of data, but I actually have 1000 columns in my data. I tried just calling the 240x1000 array into the ymon variable and it worked, but only for the first column (output is 1x7300 not 1000x7300 as I was expecting). Is there a quick way to revise this code to have it read all 1000 columns and make the shifted dataset?

Thanks again. This was very helpful.

[Kelly Kearney]

Can you clarify how your data is stored? Is it in an external file?

[Ajay Gupta]

It is not external. Its an array created by my code. (241x1000 double) trying to get to 7300x1000 double.

I think I was able to resolve by adding:

yday = zeros(7300,1000);

yday(:,:) = ymon(loc,:);

Does this make sense?

Thanks again!

Answer 2

[Sulaymon Eshkabilov]

You can import your data and easily convert them as you need using importdata() - see help DOC and convert date data into numbers using datenum() - see help DOC

Good luck

[Ajay Gupta]

Thanks!