Time series data in excel

Hi There, I have decided to use matlab instead of excel in order to make my data analysis life easier.

I have several excel files which have, for example, time (s) in column A and concentration in column B.

I've been able to import the excel file and create a plot for all the data (concentration vs time). What I need is to break up the plot into different time periods (100s,150,150....150).

For example: column A: 0-100s, 101-150s, 151-300s.. etc. and find the average concentration(column B) for that time period only so I can plot avg concentration per specified time interval.

Unfortunately it is not a simple matter of using the cell count because some data have been collected twice in 1 second, and sometimes, no data was collected for several seconds.

Any advice would be much appreciated.

Thanks in advance to all!!

Alizah

[Fangjun Jiang]

So in that case, what is your algorithm to calculate the average? Are you asking for the algorithm, or are you asking how to implement the algorithm, which you didn't disclose.

[Alizah Rahemtulla]

Thanks for your comment,

Let me clarify: Once I have the cell numbers I need, I will do something like this:

a1 = nanmean(A(1:n,10));

a2 = nanmean(A(1:n,10)); (concentration values are in column 10)

a11 = ..

Etc.. ( I will have between 8-11 depending on the file)

Then I will use something like:

P = [a1, a2, a3, a4, a5, a6, a7, a8,...]

plot(P)

I know its not very elegant, and I will have to change the number of a#'s each time (because each file is different) but I'm okay with that as I'm a matlab newbie.

I look forward to your insights

Thanks again!

Answer 1

[Fangjun Jiang]

Okay, maybe this is what you need.

1. You may want to get rid of the duplicated time and its data.

A=[1 0.1;1 0.2;2 0.3;2 0.4;3 0.5]
[trash,ind]=unique(A(:,1));
A=A(ind,:)

2. You can use logic indexing to pick out the data that meets your time criteria.

A=[(0:20)',rand(21,1)]
ind=and(5<A,A<15);
A=A(ind,:)