SAS Statistics - Logistic Regression (Module 04)

แชร์
ฝัง
  • เผยแพร่เมื่อ 1 ธ.ค. 2024

ความคิดเห็น • 6

  • @manassahoo8307
    @manassahoo8307 6 ปีที่แล้ว

    How Can I sure that which var will go to model, if I have more than 50 X variable. Where Can I get this software?
    Thanks.

  • @m051f
    @m051f 8 ปีที่แล้ว

    Where do I get the data file?

    • @zouxtion
      @zouxtion 8 ปีที่แล้ว +2

      If you google "Bank Marketing Data Set" it will appear

  • @AlexSmith-tr9hc
    @AlexSmith-tr9hc 7 ปีที่แล้ว +2

    Percent Concordant is not "percent coordinated vs. discoordinated" - LOL. You apparently have no idea what % Concordant vs. Discordant mean from the output report.
    Likewise, your Odds Ratio Estimates under the 95% Confidence Limits does not mean that they are statistically significant. These are Odds ratios within a 95% confidence band for every 1 unit of change in the independent variable. Odds ratios are asymmetric, ranging from 0 to infinity, and a value near 1 indicates no difference. A significant odds ratio is one whose confidence interval simply does not include 1.
    Likewise, the C-statistic, not the "C-score" as you indicated, estimates the probability of an individual with the outcome having a higher predicted probability than an individual without the outcome.
    While you appear to know the syntax of PROC LOGISTIC, I would suggest you take SAS Statistics I and II courses to understand what the output means from this procedure in SAS.

  • @mohsinhassan8109
    @mohsinhassan8109 4 ปีที่แล้ว

    where i can this dataset from? apparently where ever i seek it, it doesn't work

    • @josephzhang1159
      @josephzhang1159 2 ปีที่แล้ว

      Yes, me too, there is not the banking dataset as the video recorded to show. What I downloaded cannot work out