Calculating Altman Z-score
Posted: 09 April 2015 08:46 AM   [ Ignore ]
Newbie
Rank
Total Posts:  4
Joined  2015-03-30

Hi guys,

I have a gquestion regarding the Altman z-score..

I do have the calculation for this z-score, but the score cannot be calculated like this in stata:
gen T1=wcap/at
gen T2=re/at
gen T3=ebit/at
gen T4=mkvalt/lt
gen T5=sale/at
gen ALTMAN=(1.2*T1)+(1.4*T2)+(3.3*T3)+(0.6*T4)+(0.99*T5)

at = total assets
wcap = working capital
re = retained earnings
ebit = earnings before extraordinary income
mkvalt = market value of equity
sale = sales
lt = total liabilities

The formula for ALTMAN does not create the z-score in STATA
Does anyone know the path or command in STATA to calculate this z-score?

Best regards and lots of thanks

Profile
 
 
Posted: 09 April 2015 10:06 AM   [ Ignore ]   [ # 1 ]
Newbie
Rank
Total Posts:  19
Joined  2015-02-12

Coding looks fine to me… what exactly is the problem?

Here is my ALTZ stata code: gen ALTZ = 1.2*wcap/at + 1.4*re/at + 3.3*ebit/at + 0.6*(csho*prcc_f)/lt + sale/at

Profile
 
 
Posted: 10 April 2015 02:42 AM   [ Ignore ]   [ # 2 ]
Newbie
Rank
Total Posts:  4
Joined  2015-03-30
dwresearch - 09 April 2015 10:06 AM

Coding looks fine to me… what exactly is the problem?

Here is my ALTZ stata code: gen ALTZ = 1.2*wcap/at + 1.4*re/at + 3.3*ebit/at + 0.6*(csho*prcc_f)/lt + sale/at

This gives me a way to big value as being an z-score.. My mean is than -8 or something with outliers up to -68 and 75.
I assume that cannot be a correct z-score, can it?

Profile
 
 
Posted: 10 April 2015 03:01 AM   [ Ignore ]   [ # 3 ]
Newbie
Rank
Total Posts:  19
Joined  2015-02-12

I am not sure I understand correctly - the values should be firm specific and most of the values (between 1st and 99th percentile of the distribution of ALTZ) seem to be reasonable, at least in my case - I would recommend looking at the median value or maybe even trim/winsorize the distribution to cope with the outliers (or look have a closer look at the outliers - maybe these are data errors or explainable extremes)

Profile
 
 
Posted: 10 April 2015 06:56 AM   [ Ignore ]   [ # 4 ]
Administrator
Avatar
RankRankRankRank
Total Posts:  901
Joined  2011-09-19

Maybe there is a calculation error in one of the variables; what are the median values for wcap/at,re/at, etc.?

Best,

Joost

 Signature 

To reply/post new questions: Please use the group WRDS/SAS on Google Groups! http://groups.google.com/d/forum/wrdssas

Profile
 
 
Posted: 14 November 2015 10:49 AM   [ Ignore ]   [ # 5 ]
Newbie
Rank
Total Posts:  8
Joined  2014-12-16

Hello TessaM123,

I think the outliers you mention might rely on “erroneous” calculations. I don’t imply your formula is wrong, hence is Stata’s nature to regard missing variables as zero on calculations (to my knowledge at least). I suggest you include an “if” part after your formula, controlling for missing values ( !=. ). My suggestion is:

gen ALTMAN=(1.2*T1)+(1.4*T2)+(3.3*T3)+(0.6*T4)+(0.99*T5) if T1!=. & T2!=. & T3!=. & T4!=. & T5!=.

this way I think you could overcome this problem.

Antonis

 Signature 

regards,

Antonis Chantziaras

Profile