Search results

  1. S

    Calculate parameter estimate for reference level

    Hello all, I fit a complicated model in proc glimmix with 2 categorical variables (6 and 8 levels respectively) and the linear and quadratic form of a continuous variable with all interactions. The 3-way interaction is significant and I want to isolate the regression models for each combination...
  2. S

    0-3 bounded continuous dependent variable

    Hello again, I am still working on the same data and I have a question about the residual plots. Everything looks good (histogram, QQ-plot and box-plot) apart from the fitted vs residual plot (top left). I have tried many different models and I keep seeing this tilted rectangle. This model...
  3. S

    0-3 bounded continuous dependent variable

    Thank you both for the replies. There are only a handful of zeros. I can add +0.0001 to every value so I won't loose any datapoints when using log transformation. I was using glimmix to fit a beta distribution when I converted everything to (0-1) range. I will try proc mixed too. The variable...
  4. S

    0-3 bounded continuous dependent variable

    Hello all, I am working on a difficult (for my abilities) dataset. The dependent variable is continuous [0-3] bounded and was measured between 2000 to 2019 in several locations (not necessarily the same location in each year). So I have a spatial component that would like to account for. My...
  5. S

    proc import issue

    Hello, I have data with many missing observations in an excel file. Missing cells have a period ".". When using proc import, the "." is recognized as a level in subsequent analysis. However, when I paste the data directly in SAS, the "." is correctly identified as missing value. This affects...
  6. S

    Mixed model Beta distribution fit diagnostics

    Hello, I have a split-plot design and I want to test the effect of 2 factors on the disease incidence (continuous proportion). I am using Beta dist. which is appropriate for these data (bounded within 0-1). The fit statistics look OK (Pearson Chi-square/DF close to 1) Fit Statistics for...
  7. S

    proc fastclus output interpretation

    Thank you! That makes more sense now!
  8. S

    proc fastclus output interpretation

    Hello, I am using proc fastclus to perform k-means clustering. In Figure 42.2: Cluster Summary Table from the FASTCLUS Procedure, an R-square value is reported (please see link below)...
  9. S

    Mixed model LSMEANS vs ESTIMATE (BLUP)

    Thank you hlsmith for your thoughts and interest on this. I will try to email the author of the book, there is no reply to my post on SAS Communities. The std error for LSMEANS is 1.01 (p-value=0.0197) and for the BLUP is 0.83 (p-value=0.0005). The issue is that in another dataset, the...
  10. S

    Mixed model LSMEANS vs ESTIMATE (BLUP)

    I need to know why they are different. If the above BLUP estimate is the inference across the 5 locations, what is the LSMEANS then? Don't LSMEANS take into account the random effects and produce estimate of fixed effects across the 5 locations? That is what I thought. So which one should I...
  11. S

    Mixed model LSMEANS vs ESTIMATE (BLUP)

    Thank you hlsmith. I will read it thoroughly because I really need to understand the difference. It appears that LSMEAN compute the treatment effect across locations differently than the ESTIMATE (BLUP). I don't think I am doing something wrong, I follow the examples in chapter 6 in SAS for...
  12. S

    Mixed model LSMEANS vs ESTIMATE (BLUP)

    Randomized complete block design.
  13. S

    Mixed model LSMEANS vs ESTIMATE (BLUP)

    Hello, I am analyzing data from a multi-location trial (5 locations) to test the effectiveness of a treatment with 2 levels. The design is RCB with 3-4 replications in every location. I use the model below: proc mixed data=mydata; class location rep trt ; model Y=trt/ddfm=kr2 residual...
  14. S

    multi-location latin square design analysis

    Hello, I have a 5 x 5 Latin square design which is replicated 5 times within each location (same rows and columns in each location). The same design was used in 10 different locations and I was asked to perform a combined location analysis. I have done it before with other designs, such as RCB...
  15. S

    meta-analysis vs. random effect combined analysis

    I have data from 200 similar studies, all measuring the same effect of a continuous independent variable on the same continuous response. I say similar because the designs are different (split plot vs. rcbd) and the levels of the independent variable is not the same across all studies. I have...
  16. S

    permutations "partykit"

    Hello, I am new to R and I was reading about conditional decision trees. In the "party" package there is an option to select number of permutations (nresample=...). However, that is not the case with "partykit". So does it use permutations, even if it is a constant number and I can't...
  17. S

    split-split plot with continuous subplot variable

    Hello I am trying to analyze data from a split-split-plot design. The sub-plot is a continuous factor and since we suspect a non-linear relationship, the quadratic form needs to be tested as well. Factors: a-main plot-5 levels b-subplot-continuous c-sub-subplot-2 levels. To test the quadratic...
  18. S

    specify correct random effects

    Hi all, I have a CRD with 4 reps and 4 treatments (A, B, C, D). The study took place in 1 location for 3 years. I want to pool over years (so treat year as random effect). I am interested in main effects and up to 2-way interactions. So I am using the following model and random statement...
  19. S

    parameter estimate of categorical variable

    I run multiple regression with 2 continuous and 1 categorical variable (3 levels). SAS will hold the last level of the categorical variable and will not give an estimate. I know that this is the intercept. My question is how to calculate the interaction of the continuous variable with the 3rd...
  20. S

    non-mutually exclusive outcomes

    Hi, I have 4 outcomes (A B C D) that one of them is calculated as a weighted average of the first 2 (D=0.6*B + 0.4*C). Then I ran ANOVA to examine the effect of 2 factors on the 4 outcomes. My analysis was rejected because they said that D is not mutually exclusive from all other factors...