Compute probability of the reduction parameter relative to a given threshold
fecr_probability.RdComputes probability of the reduction parameter's marginal posterior density relative to a threshold.
fecr_probs(stanFit, threshold = 0.95, lessthan = TRUE, plot = TRUE, xlab, ylab, main, verbose = TRUE, ...)
Arguments
| stanFit | |
|---|---|
| threshold | numeric. The default threshold is 0.95 (95%). |
| lessthan | logical. If TRUE, the probability less than the threshold is computed. Otherwise greater or equal to the threshold is computed. Default is TRUE. |
| plot | logical. If TRUE, the posterior density of the reduction is plotted with region less than the threshold shaded. |
| xlab, ylab, main | strings. Arguments for plotting. Only used if |
| verbose | logical. If TRUE, a statement with computed probability is printed. |
| ... | additional plotting arguments |
Value
Returns a numeric value indicating the probability in percentage.
Examples
# \donttest{ ## load sample data data(epgs) ## apply individual efficacy model to the data vectors model <- fecr_stan(epgs$before, epgs$after, rawCounts = FALSE, preCF = 10, paired = TRUE, indEfficacy = TRUE)#> #> SAMPLING FOR MODEL 'indefficacy' NOW (CHAIN 1). #> Chain 1: Gradient evaluation took 3.5e-05 seconds #> Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 0.35 seconds. #> Chain 1: Iteration: 1 / 2000 [ 0%] (Warmup) #> Chain 1: Iteration: 500 / 2000 [ 25%] (Warmup) #> Chain 1: Iteration: 1000 / 2000 [ 50%] (Warmup) #> Chain 1: Iteration: 1001 / 2000 [ 50%] (Sampling) #> Chain 1: Iteration: 1500 / 2000 [ 75%] (Sampling) #> Chain 1: Iteration: 2000 / 2000 [100%] (Sampling) #> Chain 1: Elapsed Time: 1.45455 seconds (Warm-up) #> Chain 1: 0.899831 seconds (Sampling) #> Chain 1: 2.35438 seconds (Total) #> #> SAMPLING FOR MODEL 'indefficacy' NOW (CHAIN 2). #> Chain 2: Gradient evaluation took 1.6e-05 seconds #> Chain 2: 1000 transitions using 10 leapfrog steps per transition would take 0.16 seconds. #> Chain 2: Iteration: 1 / 2000 [ 0%] (Warmup) #> Chain 2: Iteration: 500 / 2000 [ 25%] (Warmup) #> Chain 2: Iteration: 1000 / 2000 [ 50%] (Warmup) #> Chain 2: Iteration: 1001 / 2000 [ 50%] (Sampling) #> Chain 2: Iteration: 1500 / 2000 [ 75%] (Sampling) #> Chain 2: Iteration: 2000 / 2000 [100%] (Sampling) #> Chain 2: Elapsed Time: 1.64621 seconds (Warm-up) #> Chain 2: 0.917111 seconds (Sampling) #> Chain 2: 2.56332 seconds (Total) #> Model: Bayesian model without zero-inflation for paired design allowing individual efficacy #> Number of Samples: 2000 #> Warm-up samples: 1000 #> Thinning: 1 #> Number of Chains 2 #> mean sd 2.5% 50% 97.5% HPDLow95 #> FECR 0.9623 0.0566 0.8083 0.9869 1.0000 0.8438 #> meanEPG.untreated 1428.6307 747.3796 511.3575 1241.6718 3265.4171 429.1522 #> meanEPG.treated 53.6706 96.2146 0.0002 16.0360 344.9586 0.0000 #> mode HPDHigh95 #> FECR 0.9966 1.0000 #> meanEPG.untreated 899.5017 3053.2711 #> meanEPG.treated 3.6331 222.9084 #> #> NOTE: there is no evidence of non-convergence since all parameters have potential scale reduction factors (Brooks and Gelman, 1998) less than 1.1.fecr_probs(model$stan.samples)#> The probability that the reduction is less than 0.95 is 25.1 %.# }
