| Class | Analysis and data manipulation command |
| Name | trend |
| Arguments | <quantitative trait> <marker locus> |
Prints Jonckheere-Terpstra trend test result for the specified marker loci versus the trait. Most suitable for SNP markers, where genotype ordering is clear. Monte-Carlo empiric P-values are produced either via gene-dropping or permutation.
Example:
>> include williamsex.in
>> kruskal adjChol ldl
------------------------------------------------
Cross-tabulation of "ldl" ... "adjChol"
------------------------------------------------
adjChol
ldl Mean SD Count
---------- ---------------------------------------
1/1 117.7917 48.3469 24
1/2 271.3684 130.6999 19
2/2 206.6667 105.9733 3
Kruskal-Wallis H= 17. df= 2 (P=0.0002)
>> set iter 50000
>> trend ace aceins
------------------------------------------------------
Nonparametric association analysis of trait "adjChol" v. "ldl"
------------------------------------------------------
Number of genotypes = 3
Number of unique trait values = 44
Number of observations = 46
Jonckheere-Terpstra Statistic = 469.5
Equalled or exceeded by = 18/50000 simulated values (0.0007)
Top simulated S:
474.0 479.0 481.0 482.0 487.0 488.5 490.0 490.5 496.5 497.5 503.5
Simulated mean (variance) = 268.4 ( 5393.3)
Asymptotic (Wald) P value = 0.0062
| << (sdt) | Up to index | >> (asp) |