Compute operation with different index (part II)

Good morning,

i've this table and i would like to build a function, where i could insert the time series as input and another input, an integer "n", that has a function of a loockback period.

example: Adj_Close(today) - Adj_Close(lookback) and i would like to apply to the all time series

example data in the attchment

[Image Analyst]

Give a small example with actual data and show what the output should look like.

[Attilio Pittelli]

the output as i wrote before should be

AdjClose(today) - AdjClose(lookback)

then

AdjClose(lookback) - AdjClose(2*lookback)

and so on

Answer 1

[Walter Roberson]

It would be easier if you would switch over to table or timetable instead of using timeseries .

filename = 'https://www.mathworks.com/matlabcentral/answers/uploaded_files/750909/AAPL.csv';
t = readtable(filename, 'VariableNamingRule', 'preserve');
tt = table2timetable(t);
ts = timeseries(tt);

ts

  timeseries

  Common Properties:
            Name: 'unnamed'
            Time: [251x1 double]
        TimeInfo: tsdata.timemetadata
            Data: [251x6 timetable]
        DataInfo: tsdata.datametadata

n = randi(10)

n = 8

result = a_function(ts, n);
result

  timeseries

  Common Properties:
            Name: 'unnamed'
            Time: [243x1 double]
        TimeInfo: tsdata.timemetadata
            Data: [243x1 double]
        DataInfo: tsdata.datametadata

function result = a_function(ts, n)
  T = ts.Time(n+1:end);
  adj = ts.Data.('Adj Close');
  result = timeseries(adj(n+1:end) - adj(1:end-n), T);
end