Search results

  1. O

    Critical F value for testing Full vs Reduced models.

    If you look at my post, I wanted the OP to come to this conclusion after seeing what results from calculating and F stat and a t-stat. I assumed this is homework so I took a harder stance :cool:
  2. O

    Critical F value for testing Full vs Reduced models.

    What are the pertinent formulae for computing F-statistics? What is the formula for testing individual coefficients with a T-test? Do you need one, both, or neither? Why?
  3. O

    Exponential smoothing models

    I agree with him on those grounds, absolutely, but that depends on the context. I think prediction is different from inferential modeling. In your case, you want practical results that equate to good predictions, right? If the toenail length of the firm's CEO allowed you to accurately predict...
  4. O

    Exponential smoothing models

    Some people in Economics believe that model assumptions shouldn't be based on how reasonable they are but how the model performs in practice. Maybe there isn't seasonality but the seasonality variable is proxying for another variable? If the end goal is predictive accuracy, then predictive...
  5. O

    Best statistical method for CFU/ml?

    Colony forming units per milliliter... if I were to take a guess... density of bacteria (or fungi) in (usually) a body fluid
  6. O

    Factor Analysis

    As far as I recall, SAS has different methods of PCA and FA since they are different methods with different goals. People often misuse one in place of the other, though, so be careful what you read on this internet!
  7. O

    Retrospective intention-to-treat analysis? what does it mean?

    Is it ITT in a retrospective study? ITT in a prospective study or RCT is a lot better because it mimics the real world situation where a physician may prescribe, but patients may end up as noncompliant in one way or another. I don't think ITT is bad. The BS to watch for is ""modified" ITT or...
  8. O

    Retrospective intention-to-treat analysis? what does it mean?

    I think these are valid points. Of course, randomization is the better approach, but attempting to account for these "imbalances" can be done but will still leave the study open to criticism. I'm just stating how it could be done. I think it is similar to a retrospective cohort study, in a way.
  9. O

    Newey-West Standard Errors

    Yeah, I'm not sure it's appropriate in an autoregressive model. Half those fancy remedies are for the econometricians who want to bulldoze through with OLS.
  10. O

    Retrospective intention-to-treat analysis? what does it mean?

    You can look at the first clinic note for when a treatment was supposed to be initiated, and you can see when that was and what the treatment was. You can follow clinic notes to see if the patient had been compliant through the period of study. (For completeness of analysis, although ITT would...
  11. O

    Logistic regression

    I'm unsure if the OP is a similar background/mindset, but my experience is that medical researchers do this a lot because they want a "firm" guideline to say if X then Y, otherwise do Z. Blood pressure guidelines, abnormal lab values, many other things where medicine dichotomizes so they can...
  12. O

    Logistic regression

    Just because the software allows you to "fit them all in the model" doesn't mean it's the right thing to put them all in the model. Having more parameters estimated relative to fewer at a given sample size runs risks of overfitting, inappropriate parameter estimates, and other problems. Sample...
  13. O

    which test for 4level categorical IV and continious IV?

    Can you provide more information on the dependent variable? How is it generated? Do people answer a questionnaire 1-5 (low-high), for example?
  14. O

    Interpreting Hosmer-Lemeshow Goodness of Fit?

    If I recall, the null is good fit. Lower p-values suggest more evidence to contradict "good fit". Larger p-values just mean you don't have enough evidence of "bad" fit.
  15. O

    High statistical significance with low R squared coefficient

    Right, but the OP is presenting a Frequentist statitstic, and encouraging the interpretation in the Bayesian framework contributes to the poor literacy we see today and the illusion that a p-value of .08 means there's an 8% chance the null is true or that a "mistake" has been committed. You and...
  16. O

    High statistical significance with low R squared coefficient

    So you're trying to move from a Frequentist interpretation to a Bayesian one. I think your argument, in the Frequentist framework, is flawed in that the null is either true or it is not, "probability" 1 or 0. The p-value doesn't change that in the slightest. It also doesn't imply that [small...
  17. O

    High statistical significance with low R squared coefficient

    As far as I'm aware, the easiest way would be introducing biases that readily occur in practice. If you don't want to, you could certainly repeat the experiment as many times as you want to keep interpreting p-values and end up making many different claims regarding the probability of Ho based...
  18. O

    High statistical significance with low R squared coefficient

    Sure, if you want a real experiment versus a simulation, either can be done. Failure to randomize people into groups in a way that leads to bias away from the null, or introducing some other form of bias that typically occurs in studies and you can get a p-value less than alpha, but the null is...
  19. O

    High statistical significance with low R squared coefficient

    This is what I was saying is incorrect. The p-value doesn't tell you anything regarding the probability the null hypothesis is true or false. I can design an experiment where the null hypothesis is true but where the p-value is very low. This illustrates why it's incorrect to say the p-value...
  20. O

    High statistical significance with low R squared coefficient

    The bold part isn't accurate. A small p-value for the case you provided would indicate that it's improbable to see a coefficient at least as contradictory to the null hypothesis, IF what we saw is entirely due to chance (null is true). In other words, p-values don't tell you any sort of...