Recent content by hlsmith

  1. hlsmith

    Anyone can help me with choosing classes?

    OR = ? So you took every course with linear algebra or matrix in it? I would take the applied regression courses and other applied courses, you would be surprised how many questions you can answer using regression and they are face value interpretable. You should look to your school's...
  2. hlsmith

    Anyone can help me with choosing classes?

    I believe many stats programs requires between 2-3 semesters of calculus and 1-2 in linear algebra. If you want to be competitive, I would ensure you take the higher range of both of those and do well. As for other courses, are you hoping to apply ML to a certain field - if so I would keep that...
  3. hlsmith

    How are the p-values calculated in a Ordered Logistic Regression

    @GretaGarbo - I was referring to this site: http://talkstats.com/ - and if you are going to just tell them to use Google. It was kind of a joke on my part and a response to your directness.
  4. hlsmith

    How are the p-values calculated in a Ordered Logistic Regression

    @GretaGarbo - then why should this site exist? @statsnooooob - I have run many logistics, but very surprisingly never an ordered logistic regression. I would imagine it is very comparable to a traditional logistic reg, but has interval groups ordered, compared to a reference level. Sorry I...
  5. hlsmith

    Is there a test for omitted variable bias?

    If there are two groups of kids, one gets a computer, one doesnt. Later you see which group was more likely to pass a test. Well was it the computer or that one group had a disproportionate high number of honors students. So the computer could have helped or it was the kids would have done...
  6. hlsmith

    Partial reggression plots

    Well you can use LOESS, but since you mentioned partial, GAMs allow you to throw in covariates. I run mostly logistic reg, so here is an example for that: https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_gam_sect027.htm
  7. hlsmith

    P values versus slope

    Yeah, but in every instance you are making a decision based on a binary rule and why does 0.05 matter instead of 0.1 or if you have a huge dataset, 0.001? They are just making a bunch of incremental hypotheses. That would probably make a good paper, how many statistical test get ran when...
  8. hlsmith

    regression degrees of freedom

    When you do a PROC CONTENTS, how is it defining these variables? If they are numbers, you gotta let the program know or if using PROC REG use created dummies - I know you know how to do that. I remember you working with me once on an option in PROC REG and using dummies.
  9. hlsmith

    Competing risk COX in SPSS

    Yeah, I primarily use SAS for Cox, but I am truly sure it can be done in R. Well in 1995 it couldn't do it: https://groups.google.com/forum/#!topic/comp.soft-sys.spss/oQwg37MSTfo I wonder if there is a SPSS Help Forum you can consult or just give the company a call.
  10. hlsmith

    Is there a test for omitted variable bias?

    If you do not randomize the exposure and also control for extraneous factors you can never truly know if an omitted variable may be impacting the outcome. With enough evidence and causal assumptions, you can presume a causal relationship, but you just need to posit a confounder could be...
  11. hlsmith

    regression degrees of freedom

    So if I remember proc reg doesn't have the class statement, so you have to dummy code them right?! Which variables should be treated as categorical?
  12. hlsmith

    Competing risk COX in SPSS

    Not a SPSS user, but I would imagine it should be available? Sorry I don't have specifics, just acknowledging you post - not a tremendous amount of SPSS user currently on this forum - but keep the posts coming.
  13. hlsmith

    Mixed model an linear model gave the same results

    Typically if you think you have a multilevel model, but controlling for groups does not explain any additional variability (ICC), and its not a power issue (sample size), people will revert back to a OLS model and use sandwich SE estimators for simplicity and sandwich for staving off type I errors.
  14. hlsmith

    Partial reggression plots

    When I test a continuous variable for linearity in the model, I usually just use a SAS gam model (I think gampl, etc.) and look at the plotted relationship and how many if any knots there are on it.
  15. hlsmith

    Greetings

    Welcome to the forum Grant. Not sure why your music is pretentious- its just music. We look forward to your questions and posts!