How to aggregate data and calculate its standard deviation

Hi

I have two matrix with 1) day 2) travel time (sec). Where there are several observations of travel times per day. And date for more than two years.

day .......Traveltime

1 .............220

1 .............230

1 .............240

2 .............210

2 .............210

...etc

How can I aggregate the data_, so that I will have only one observation (average travel time) per day, and the standard deviation for each day.

So that the output, will be something like this:

day.......average_time.......Std_dev

1............230.............10

2............210.............0

I would appreciate if you can help me.

[per isakson]

Firstly, a couple of questions:

• The input data is that a text file? Or double array? Or a cell array?
• " are several observations of travel times per day" does that mean that the length of the lines/row varies?

[Azzi Abdelmalek]

Duplicate

Answer 1

[dpb]

I'll presume you have two vectors d and t; if so you can make a matrix as

MATl
>> m = [d t]
m =
   1   220
   1   230
   1   240
   2   210
   2   210
>> [u,~,c] = unique(m(:,1));   % get the unique days, locations
>> [u,accumarray(c,m(:,2),[],@mean),accumarray(c,m(:,2),[],@std)]
ans =
   1   230    10
   2   210     0
>>

Or, of course, you can use the two vectors directly w/o the concatenations...

MATL
>> [u,~,c] = unique(d);
>> [u,accumarray(c,t,[],@mean),accumarray(c,t,[],@std)]
ans =
   1   230    10
   2   210     0
>>

Why didn't just use them first time is beyond me... :)