About Crubal

Crubal · ‎06-06-2017

Hi Rick, I read through your blogs, those are extermely helpful. Thank you! For this particular question, My initial probability distributon is like this : This is the actrual case, however, I want to adust the probability ofr some values: For example, in the above figure. I want the Probability of 'No-show' = 6 be smaller than 'No-show' = 5; and 'No-show' = 7 be in the middle of 'No-show' = 6 & 9. Which looks like a Normal or Poisson Distribution, and finally returns probability for each value of 'No-show'. My current approach is to obtain mean for this group, and generate Poisson Distribution according to it. Like the following (Not exact the same data, but similar case) : So that in this way, the probability is increasing before the mean and then decreasing after it. Is there any other approach? Thank you!

Crubal · ‎06-06-2017

Hi Ksharp, Sorry for my late response. The initial probability table looks like this: For example, the above figure. I want the Probability of 'No-show' = 6 be smaller than 'No-show' = 5; and 'No-show' = 7 be in the middle of 'No-show' = 6 & 9. Which looks like a Normal or Poisson Distribution, and finally returns probability for each value of 'No-show'. My current approach is to obtain mean for this group, and generate Poisson Distribution according to it. Like the following (Not exact the same data, but similar case) In this way, the probability keeps increasing before the mean value, and then keeps decreasing after that. Is there any other approach? Thank you!

Crubal · ‎06-01-2017

Thanks ballardw, I just removed 'output' in the data step as you suggested, and it returned one value for each date. This is what I need. Thanks!

Crubal · ‎06-01-2017

Hi ballardw, Thanks for your reply! As I used max function, for example 'max((1-i), 0)'. It will only return 0 for i = 1 But I did get 10 values of 'Q_1' for each Date. Maybe because I set i =1 to 10? Thank you!

Crubal · ‎06-01-2017

Hi, My input data c1 is as following: Date | P1 | P2 | Mean ----------------------------------------- 01JUN2017 | 200 | 300 | 3.2 ----------------------------------------- 02JUN2017 | 230 | 320 | 2.7 ----------------------------------------- 03JUN2017 | 250 | 340 | 3.7 ----------------------------------------- For each row, I would like to calculate Q_1, which based on the following function (pseudo code): Do i = 1 to 10: Q_1 = sum(P1 * pdf('Poisson', 1, Mean) * max((i-1),0) + P2 * pdf('Poisson', 1, Mean) * max((1-i),0)) so that output table c2 will have another column called 'Q_1' for each date. My initial code: Data c2; Set c1; do i = 1 to 10; Q_1 = sum(P1 * pdf('Poisson', 1, Mean) * max((i-1),0) + P2 * pdf('Poisson', 1, Mean) * max((1-i),0)); output; end; drop i; Run; But it returns 10 values of 'Q_1' for each date. How could I write it in data step? Thank you!

Crubal · ‎05-17-2017

Hi DW, I am sorry I have got additional question. After using 'Proc Autoreg' procedure under the prior code, and obtained 'yhat' for predicted value. As variable 'time' I set is used applied among passed dates. And this variable has date from '01MAY2013' to '30APR2018'. If I would like to apply the model in forecasting future dates, which is '01MAY2017' to '30APR2018', how could I write the code? Thank you so much! Initial Code: Proc Autoreg Data = Work.data1; ods output FitSummary=fitstats; /* added statement */ Model Y = time / nlag = 3 method = ml backstep; by A B C; output out = Work.Result_1 p = yhat pm = trendhat; where time between '01MAY2013'd and '30APR2017'd; Run; proc print data=fitstats; run;

Crubal · ‎05-17-2017

Thank you so much DW!

Crubal · ‎05-17-2017

Thank you DW, this is extremely helpful! I removed 'Backstep', However, I looked at the 'FitStats' table, for each combination of 'A', 'B' & 'C', there are two 'MSE' &' MAE' for column 'Label1'. which one shall I use? Thank you!

Crubal · ‎05-16-2017

Hi, I used 'Proc Autoreg' procedure to predict a time trend data and have got a question related to it. My code: Proc Autoreg Data = Work.data1; Model Y = time / nlag = 3 method = ml backstep; by A B C; output out = Work.Result_1 p = yhat pm = trendhat; Run; Itr runs successfully, however, summary statistics shows in 'Result' page which is printed, how could I obtain its related MSE & MAE in an output dataset? Thank you!

Crubal · ‎05-12-2017

Hi DW, Thanks! I will take a look at the link. By the way, using 'Proc ESM' and model = damptrend, future demand always gets the same, is it always the case from 'Exponential Smoothing' point of view? Thank you! Chenxi

Crubal · ‎05-11-2017

Hi DW, Thank you, as you mentioned it works thanks for your recommendation! I did try on 'Proc Forecast' functon like below using Exponential Smoothing, will that proc forecast data=factory interval=day method = Expo lead = 365 out = fcst outest=est by factory item; id date; var demand; where Date between '01MAY2013'd and '30APR2016'd; run; Will that result the same? And by the way, is there any change that I could make for the 'Method = ' part so that I could use empirical mean to represent future date demand? Thanks!

Crubal · ‎05-11-2017

Hi, Thank you! It is pretty clear and I appreciate. If I would like to forecast 1-year daily data (assuming weekend has demand), shall I make some changes to theparameter of your code? (Back, Lead, Interval) proc esm data=factory back=365 lead=730 outfor=fcst outest=est outstat=stat print=all plot=modelforecasts; by factory item; id date interval=day; forecast demand / model=damptrend; run; It seems error message shows: Back = 365 is greater than the number of Time Series Value Thanks

Crubal · ‎05-10-2017

Hi, I have the following dataset (Table A) Factory | Item | Date | Actual_Demand ------------------------------------------------------------- A | 1 |2013-05-01 | 21 ------------------------------------------------------------- A | 1 |2013-05-02 | 19 ...... .... ................. ..... ------------------------------------------------------------- A | 1 |2017-04-30 | 15 ...... .... ................. ...... So for each combination of 'Factory' and 'Item', we have got actual demand for 4 years. And I want to forecast future demand by 'Factory' and 'Item' level. Use the first 3-year data as training, and the last year data for testing. And I want to try 'exponential smoothing' method. How could I write the code? Will 'Proc Esm' help? Thank you!

Crubal · ‎05-10-2017

Hi, I have a dataset with name 'Data_1' with four columns: 'Factory', 'Item', 'Date', 'Actual_Demand' & 'Prediction_Demand'. I want to calculate MAPE and MSE that 'Prediction' via 'Actual' group by 'Factory' & 'Item' across 'Date'. And my code is: Proc Sql; Create Table Work.Result_1 as select distinct Factory, Item, avg(abs((Actual_Demand - Prediction_Demand)/Actual_Demand)) as MAPE, avg((Prediction_Demand - Actual_Demand)**2) as MSE from Work.Data_1 group by 1, 2; Quit; So that I could obtain a new table with two new columns. Is the formula I have in the code correct to calculate 'MAPE' & 'MSE'? Thank you very much!

Crubal · ‎05-10-2017

Hi, I have a dataset with name 'Data_1' with four columns: 'Factory', 'Item', 'Date', 'Actual_Demand' & 'Prediction_Demand'. I want to calculate MAPE and MSE that 'Prediction' via 'Actual' group by 'Factory' & 'Item' across 'Date'. And my code is: Proc Sql; Create Table Work.Result_1 as select distinct Factory, Item, avg(abs((Actual_Demand - Prediction_Demand)/Actual_Demand)) as MAPE, avg((Prediction_Demand - Actual_Demand)**2) as MSE from Work.Data_1 group by 1, 2; Quit; So that I could obtain a new table with two new columns. Is the formula I have in the code correct to calculate 'MAPE' & 'MSE'? Thank you very much!

Online Status	Offline
Date Last Visited	‎10-22-2018 11:14 AM

Re: Add a label column based on value difference

Re: Add a label column based on value difference

Add a label column based on value difference

Re: Does proc hpclus have 'by' function?

Does proc hpclus have 'by' function?

Re: How to delete top and bottom value within a group

Re: How to delete top and bottom value within a group

Re: How to delete top and bottom value within a group

How to delete top and bottom value within a group

Re: sas contains function cannot work properly

Re: Add a label column based on value difference

Re: sas contains function cannot work properly

Re: Date and Time Issue

Re: proc sql insert a row

Re: Add a check point to the entire procedure

Re: According to a group of data, how to simulate a distribution

Re: According to a group of data, how to simulate a distribution

Re: do loop sum in data step

Re: do loop sum in data step

do loop sum in data step

Re: How to output MSE / MAE information into an output dataset?

Re: How to output MSE / MAE information into an output dataset?

Re: How to output MSE / MAE information into an output dataset?

How to output MSE / MAE information into an output dataset?

Re: SAS Exponential Smoothing

Re: SAS Exponential Smoothing

Re: SAS Exponential Smoothing

SAS Exponential Smoothing

MSE & MAPE Function correct?

How to calculate MAPE & MSE according to two columns of data