Dear Community,

The code below is designed to pass data to Python, calculate internal rates of return and return the IRR to SPSS.  When I use the SetValueNumeric method of spss.Cursor, the new variable is passed to SPSS as all missing values.  I can't figure out why.  The correct values are in Python and I can get them into SPSS by printing them to a text file and reading the text file into SPSS.  The program and output is copied below.  I've also attached a copy of the output file.  I'm using SPSS version 29.

I appreciate your help!

Best regards,

Steve O'Byrne

get file = 'data file for dollar weighted returns.sav'.

* select if (gvkey eq '12124').

* list variables = gvkey year horizon basemthnum endmthnum.

set printback on mprint on.

BEGIN PROGRAM PYTHON.

import spss

import numpy_financial as npf

spss.Submit("Frequencies variables = horizon.") # this passes a command to SPSS without transfering any data to Python

dataCursor = spss.Cursor(accessType='w')

dataCursor.AllocNewVarsBuffer(8) # needed to save one numeric variable

allCases = dataCursor.fetchall() # allCases is a list of tuples; each tuple has 542 (= 5 + 537) values

print("Number of cases is ",len(allCases))

gvkey = []

year = []

horizon = []

basemthnum = []

endmthnum = []

irrCases = []

irr_calc = []

for i in range(len(allCases)):

    gvkey.append(allCases[i][0])

    year.append(allCases[i][1])

    horizon.append(allCases[i][2])

    basemthnum.append(allCases[i][3])

    endmthnum.append(allCases[i][4])

for i in range(len(allCases)):

    temp_list = []

    irrCases.append(temp_list)

print("Length of irrCases is ",len(irrCases))

for i in range(len(allCases)):   

    for j in range(len(allCases[i])):

         if (j > (basemthnum[i]+3) and j < (endmthnum[i]+5)):

              irrCases[i].append(allCases[i][j])

for i in range(len(irrCases)):

    irr_calc.append(round(npf.irr(irrCases[i]),6))

dataCursor.SetVarNameAndType(['mthly_irr'],[0])

dataCursor.SetVarFormat('mthly_irr',5,9,6)

dataCursor.CommitDictionary()

for i in range(len(irrCases)):   

    dataCursor.fetchone()

    dataCursor.SetValueNumeric('mthly_irr',irr_calc[i])

    dataCursor.CommitCase() # this code creates variable but all values are missing

dataCursor.close()                    

outfile = open("c:/execcomp/2023_data/dollar weighted returns.txt","w")

for i in range(len(irrCases)):

    print(gvkey[i],year[i],horizon[i],basemthnum[i],endmthnum[i],len(irrCases[i]),irr_calc[i],sep=',',file = outfile)

outfile.close()

END PROGRAM.

set printback = on mprint = off.

frequencies variables = mthly_irr.

DATA LIST file = 'c:\execcomp\2023_data\dollar weighted returns.txt' FREE (',')/

    gvkey (a7) year horizon basemthnum endmthnum dwrtr_cases mthly_irr.

compute hzn_irr = ((1 + mthly_irr)**(endmthnum - basemthnum)) - 1.

variable labels hzn_irr 'Dollar Weighted Return'.

formats mthly_irr (f7.5) hzn_irr (f8.4).

temporary.

select if (gvkey eq '12124').

list variables = gvkey mthly_irr hzn_irr/cases = 100.

frequencies variables = hzn_irr/format = notable/percentiles = 10 25 50 75 90/statistics = mean maximum minimum.

sort cases gvkey year horizon.

save outfile = 'dollar weighted returns by gvkey year horizon.sav'.

Stephen F. O'Byrne

President

Shareholder Value Advisors Inc

21 Bonnie Way

Larchmont, NY 10538

914-833-5891 cell 914-672-1646

Dear Community,

The code below is designed to pass data to Python, calculate internal rates of return and return the IRR to SPSS.  When I use the SetValueNumeric method of spss.Cursor, the new variable is passed to SPSS as all missing values.  I can't figure out why.  The correct values are in Python and I can get them into SPSS by printing them to a text file and reading the text file into SPSS.  The program and output is copied below.  I've also attached a copy of the output file.  I'm using SPSS version 29.

I appreciate your help!

Best regards,

Steve O'Byrne

get file = 'data file for dollar weighted returns.sav'.

* select if (gvkey eq '12124').

* list variables = gvkey year horizon basemthnum endmthnum.

set printback on mprint on.

BEGIN PROGRAM PYTHON.

import spss

import numpy_financial as npf

spss.Submit("Frequencies variables = horizon.") # this passes a command to SPSS without transfering any data to Python

dataCursor = spss.Cursor(accessType='w')

dataCursor.AllocNewVarsBuffer(8) # needed to save one numeric variable

allCases = dataCursor.fetchall() # allCases is a list of tuples; each tuple has 542 (= 5 + 537) values

print("Number of cases is ",len(allCases))

gvkey = []

year = []

horizon = []

basemthnum = []

endmthnum = []

irrCases = []

irr_calc = []

for i in range(len(allCases)):

    gvkey.append(allCases[i][0])

    year.append(allCases[i][1])

    horizon.append(allCases[i][2])

    basemthnum.append(allCases[i][3])

    endmthnum.append(allCases[i][4])

for i in range(len(allCases)):

    temp_list = []

    irrCases.append(temp_list)

print("Length of irrCases is ",len(irrCases))

for i in range(len(allCases)):   

    for j in range(len(allCases[i])):

         if (j > (basemthnum[i]+3) and j < (endmthnum[i]+5)):

              irrCases[i].append(allCases[i][j])

for i in range(len(irrCases)):

    irr_calc.append(round(npf.irr(irrCases[i]),6))

dataCursor.SetVarNameAndType(['mthly_irr'],[0])

dataCursor.SetVarFormat('mthly_irr',5,9,6)

dataCursor.CommitDictionary()

for i in range(len(irrCases)):   

    dataCursor.fetchone()

    dataCursor.SetValueNumeric('mthly_irr',irr_calc[i])

    dataCursor.CommitCase() # this code creates variable but all values are missing

dataCursor.close()                    

outfile = open("c:/execcomp/2023_data/dollar weighted returns.txt","w")

for i in range(len(irrCases)):

    print(gvkey[i],year[i],horizon[i],basemthnum[i],endmthnum[i],len(irrCases[i]),irr_calc[i],sep=',',file = outfile)

outfile.close()

END PROGRAM.

set printback = on mprint = off.

frequencies variables = mthly_irr.

DATA LIST file = 'c:\execcomp\2023_data\dollar weighted returns.txt' FREE (',')/

    gvkey (a7) year horizon basemthnum endmthnum dwrtr_cases mthly_irr.

compute hzn_irr = ((1 + mthly_irr)**(endmthnum - basemthnum)) - 1.

variable labels hzn_irr 'Dollar Weighted Return'.

formats mthly_irr (f7.5) hzn_irr (f8.4).

temporary.

select if (gvkey eq '12124').

list variables = gvkey mthly_irr hzn_irr/cases = 100.

frequencies variables = hzn_irr/format = notable/percentiles = 10 25 50 75 90/statistics = mean maximum minimum.

sort cases gvkey year horizon.

save outfile = 'dollar weighted returns by gvkey year horizon.sav'.

Stephen F. O'Byrne

President

Shareholder Value Advisors Inc

21 Bonnie Way

Larchmont, NY 10538

914-833-5891 cell 914-672-1646