The goal of bpnreg is to fit Bayesian projected normal regression models for circular data.
You can install bpnreg from github with:
This is a basic example which shows you how to run a Bayesian projected normal regression model:
library(bpnreg)
bpnr(Phaserad ~ Cond + AvAmp, Motor)
#> Projected Normal Regression
#>
#> Model
#>
#> Call:
#> bpnr(pred.I = Phaserad ~ Cond + AvAmp, data = Motor)
#>
#> MCMC:
#> iterations = 1000
#> burn-in = 1
#> lag = 1
#>
#> Model Fit:
#> Statistic Parameters
#> lppd -56.98665 8.000000
#> DIC 130.03264 7.978867
#> DIC.alt 129.00526 7.465178
#> WAIC 130.13423 8.080461
#> WAIC2 131.93160 8.979148
#>
#>
#> Linear Coefficients
#>
#> Component I:
#> mean mode sd LB HPD UB HPD
#> (Intercept) 1.38838877 1.430450408 0.44851825 0.54791575 2.21283384
#> Condsemi.imp -0.55387711 -0.586686873 0.61704234 -1.66603600 0.62531778
#> Condimp -0.64634612 -0.671047696 0.67977534 -1.86378099 0.74237047
#> AvAmp -0.01081638 -0.007612952 0.01192791 -0.03374693 0.01254156
#>
#> Component II:
#> mean mode sd LB HPD UB HPD
#> (Intercept) 1.43186794 1.34887463 0.42859821 0.61193193 2.26798239
#> Condsemi.imp -1.21413507 -1.31468438 0.58965151 -2.29088651 -0.01454310
#> Condimp -0.97439306 -1.21569705 0.63152408 -2.20567414 0.22262240
#> AvAmp -0.01174821 -0.01165855 0.01121201 -0.03183777 0.01192664
#>
#>
#> Circular Coefficients
#>
#> Continuous variables:
#> mean ax mode ax sd ax LB ax UB ax
#> 91.20303 70.15309 131.17655 -104.32763 313.05562
#>
#> mean ac mode ac sd ac LB ac UB ac
#> 0.8709000 2.0828734 1.3282028 -0.8151373 2.5196304
#>
#> mean bc mode bc sd bc LB bc UB bc
#> -0.004133268 0.009906482 0.037287197 -0.035000622 0.025243901
#>
#> mean AS mode AS sd AS LB AS UB AS
#> -0.015613031 -0.006036166 0.323371494 -0.205164040 0.106346969
#>
#> mean SAM mode SAM sd SAM LB SAM UB SAM
#> 0.120674972 -0.008383957 3.571831737 -0.192199588 0.240442932
#>
#> mean SSDO mode SSDO sd SSDO LB SSSO UB SSDO
#> 0.02983466 -2.03153425 2.07108578 -2.76494006 2.78617593
#>
#> Categorical variables:
#>
#> Means:
#> mean mode sd LB UB
#> (Intercept) 0.8036304 0.7090624 0.1980157 0.4295190 1.180707
#> Condsemi.imp 0.2449676 0.3431420 0.4128835 -0.5628898 1.077003
#> Condimp 0.5505015 0.6278918 0.4600773 -0.4725950 1.344139
#> Condsemi.impCondimp -1.2906544 -1.6598716 1.0721323 3.0402344 1.082592
#>
#> Differences:
#> mean mode sd LB UB
#> Condsemi.imp 0.5601076 0.50581982 0.4810523 -0.3500191 1.504551
#> Condimp 0.2564314 0.09884444 0.5388666 -0.8239476 1.297076
#> Condsemi.impCondimp 2.1797273 2.58905512 1.0137416 -0.4643209 3.889132