calculate sum month and annual

hi, i want to calculate profit month and annual

T1 is a vector with date from 1/1/2008 to 05/08/2023 (last rilevation)

T2 is a vector with profit

i need to calculare sum of month (gen 2008..feb 2008..) and annual (2008-2009-2010 ....)

class(pT2)

double

class(T1)

ans =

'datetime'

Answer 1

[dpb]

Read up on retime and timetable which lead into readtable and table2timetable

All are "modern" MATLAB tools; the time-honored and original way would have been with accumarray

[piero]

>> retime(t,'daily','sum')

Incorrect number or types of inputs or outputs for function 'retime'.

I'm sorry but without a practical example on my code I don't know how to use them

[dpb]

You didn't show us your code, remember? :)

But, the doc for retime says the first argument must be a timetable and I gave you the others for how to get there...try

load Export_T1.mat
whos

  Name           Size             Bytes  Class    Attributes

  T_          5710x2             663961  table              
  ans            1x38                76  char               
  cmdout         1x33                66  char               

head(T_)

         T1               T2     
    _____________    ____________

    "01-Jan-2008"    "0"         
    "02-Jan-2008"    "0"         
    "03-Jan-2008"    "0"         
    "04-Jan-2008"    "0"         
    "05-Jan-2008"    "0"         
    "06-Jan-2008"    "-49640"    
    "07-Jan-2008"    "137800"    
    "08-Jan-2008"    "-107331.75"

Ewwwww....everything is a string, not native date types...gotta' fix that first; that's not what your original post says the variable types are/were...

T_.Properties.VariableNames={'Date','Profit'};  % make more meaningful names
T_=convertvars(T_,{'Date'},'datetime');         % convert types
T_=convertvars(T_,{'Profit'},'double');
ttT=table2timetable(T_);                        % now convert to timetable
clear T_                                        % no sense keeping two copies same thing in memory
format bank, format compact                     % "profit" implies currency format
head(ttT)                                       % see what we gots now...

       Date          Profit  
    ___________    __________
    01-Jan-2008          0.00
    02-Jan-2008          0.00
    03-Jan-2008          0.00
    04-Jan-2008          0.00
    05-Jan-2008          0.00
    06-Jan-2008     -49640.00
    07-Jan-2008     137800.00
    08-Jan-2008    -107331.75

pByYear=retime(ttT,'yearly',@sum);
pByYear.Date=dateshift(pByYear.Date,'end','year')

pByYear = 16×1 timetable
       Date          Profit  
    ___________    __________
    31-Dec-2008    1869053.06
    31-Dec-2009       9218.00
    31-Dec-2010     776858.50
    31-Dec-2011       4170.25
    31-Dec-2012      53796.25
    31-Dec-2013      71171.00
    31-Dec-2014      81188.00
    31-Dec-2015      25270.00
    31-Dec-2016      19488.75
    31-Dec-2017      12759.50
    31-Dec-2018      30552.25
    31-Dec-2019      33090.50
    31-Dec-2020      61167.00
    31-Dec-2021      52313.25
    31-Dec-2022      61586.50
    31-Dec-2023       6749.50

pByYear.Date.Format ='yyyy'

pByYear = 16×1 timetable
    Date      Profit  
    ____    __________
    2008    1869053.06
    2009       9218.00
    2010     776858.50
    2011       4170.25
    2012      53796.25
    2013      71171.00
    2014      81188.00
    2015      25270.00
    2016      19488.75
    2017      12759.50
    2018      30552.25
    2019      33090.50
    2020      61167.00
    2021      52313.25
    2022      61586.50
    2023       6749.50

Monthly shouldn't be hard to guess! :)

@doc:dateshift

[piero]

thanks..great lesson on how to fix the problem

[dpb]

The way forward is to read the doc syntax and examples carefully and adapt to your particular case paying particular attention to requirements on input data types, etc., ...

The link at the top of the doc page for each function that is titled "Functions" is particularly helpful when you're just learning; it will provide a compendium of all the related functions of a given area; here for a table or timetable besides just the one you happen to be looking at. Those and their brief functionality are a gold mine of hints to accomplishg a goal. Never ignore the "See Also" and, if provided "More About" sections, either. The doc is woefully underappreciated and underused...

[piero]

all correct...if you understand english well :D

[piero]

last question please..

i use retime(ttT,'monthly',@sum) and i get this:

01-Apr-2023	-1099.5
01-May-2023	-2765.75
01-Jun-2023	1521.75
01-Jul-2023	-2276
01-Aug-2023	0

it's correct but i want to visualize that:

30-apr-2023 ...(sum 1 april to 30 april 2023)

31-may-2023...(sum 1 may to 31 may 2023)

ecc

and

i use retime(ttT,'yearly',@sum) and i get this:

01-Jan-2019	33090.5
01-Jan-2020	61167
01-Jan-2021	52313.25
01-Jan-2022	61586.5
01-Jan-2023	6749.5

but i want to visualize:

2022 ---(sum from gen 2021 to de 2021)

2023 ...(sum from gen 2022 to de 2022)

[dpb]

See dateshift -- and update to previous example...

And, somehow an earlier comment I posted seems to have gotten lost -- I forget now just what I included but I did say I think your English is quite good despite your doubts. :)

[piero]

i'm using google translate :D

[Image Analyst]

It's weird that Google translate does not know that "i" and "i'm" should always be capitalized in English. And they should always capitalize their name, as with all proper names, and the first letter of each sentence.

Answer 2

[Image Analyst]

Some functions for you to look at splitapply, findgroups, grpstats, groupsummary

[dpb]

To that list add rowfun. And, while at familiarizing, then varfun is closely related but for slightly different problem type.