On 17-12-2012, at 08:45, Agnes Ayang wrote: > Hello r helpers! Below is the whole coding for my programme. Before proceed > more further, let me explain for you. First of all, I need to compute trimmed > mean. Till that step is ok. Then I need to compute ssdw which is sum of > square deviation. If I do equal trimming at both tail of distribution that I > chose, I will use the first ssd formulae which is "a". But if I am doing > unequal trimming, then I will be using "b". My problem is I dont know how to > come out with a value of ssdw. For example, if the value of gw1=gw2 (which > mean equal trimming), the result will come out as "a" not the value of ssdw > when equal trimming is done. I f anyone could with this, I really appreciate > it. I need the value of ssdw so that I can proceed with computing the F test. > Thanks. > > Regards, > Hyo Min > UPM Malaysia > > > n1<-15 > miu<-0 > sd1<-1 > > data1<-rnorm(n1,miu,sd1) > > > ## data transformation as G=0.5 and h=0.5 (data is normal) > G<-0.5 > h<-0.5 > w<-((exp(G*data1)-1)/G)*(exp((h*data1^2)/2)) > > group1<-sort(w) > > > g<-0.15 ## proportion of trimming > > ## group 1 > uw<-floor(0.2*n1) > low1<-mean(group1[1:uw]) > uppw1<-mean(group1[(n1-uw+1):n1]) > UWx<-uppw1-low1 > > vw<-floor(0.5*n1) > low2<-mean(group1[1:vw]) > uppw2<-mean(group1[(n1-vw+1):n1]) > LWx<-uppw2-low2 > > gam.low<-g*(UWx/(UWx+LWx)) > gam.upp<-g-gam.low > > gw1<-floor(n1*gam.low) > gw2<-floor(n1*gam.upp) > hw<-n1-gw1-gw2 > trim.mean1<-1/hw*(sum(group1[(gw1+1):(n1-gw2)])) > > a<-((gw1+1)*((group1[gw1+1]-trim.mean1)^2))+sum(group1[(gw1+2):(n1-gw1-1)]-trim.mean1)^2+((gw1+1)*(group1[n1-gw1]-trim.mean1)^2) > b<-(((gw1+1)*((group1[gw1+1]-trim.mean1)^2))+sum(group1[(gw1+2):(n1-gw2-1)]-trim.mean1)^2+((gw2+1)*(group1[n1-gw2]-trim.mean1)^2)-(gw1*(group1[gw1+1]-trim.mean1)+gw2*(group1[n1-gw2]-trim.mean1)^2))/n1 > ssdw<-ifelse(gw1==gw2, "a", "b")
Don't use the quotes: you want the variable or object a not the character string "a". ssdw<-ifelse(gw1==gw2, a, b) Since gw1 and gw2 are scalars you can dispense with the ifelse. This will do as well ssdw <- if(gw1==gw2) a else b or if(gw1==gw2) ssdw <- a else ssdw <- b Berend > ssdw > > ## calculate Ft test statistic > H<-hw > C<-1 > A<-hw*trim.mean1 > xbar<-A/H > > X<-hw*(trim.mean1-xbar)^2 > Y<-ssdw > df1<-1 > df2<-H-C > > upper<-X/df1 > lower<-Y/df2 > Ft<-upper/lower > Ft > pv<-1-pf(Ft,df1,df2) > [[alternative HTML version deleted]] > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
