;+ ;FUNCTION average_hist(d,x [,STDEV=stdev]) ;returns the average of d binned according to x ;See "HISTBINS" for a description of keywords ;USAGE: ;assuming: ; x = randomu(seed,1000)*10 ; y = 10-.1*x^2 + randomn(seed,1000) ; d = y ; avg = average_hist(d,x,xbins=xc) ; avg = average_hist(d,x,xbins=xc,range=[2,8],binsize=.25) ; plot,x,y,psym=3 ; oplot,xc,avg,psym=-4 ;NOTE: d can be an array of structures: ; d=replicate({x:0.,y:0.},1000) ; d.x = x ; d.y = y ; plot,d.x,d.y,psym=3 ; avg = average_hist(d,d.x) ; oplot,avg.x,avg.y,psym=-4 ;- function average_hist,a,x,stdev=std,log=log, $ range=range,binsize=binsize,nbins=nbins,xbins=xbins,$ binval=bins, minimum=minimum,shift=shft,nan=rnan,$ histogram=h,reverse=ri if keyword_set(x) then begin if n_elements(x) ne n_elements(a) then message,'Inputs must have same number of elements' bins = histbins(x,xbins,log=log,range=range,nbins=nbins,binsize=binsize,/retbins,shift=shft) endif if keyword_set(bins) then begin if n_elements(bins) ne n_elements(a) then message,'BINS must be same size as input array' if not keyword_set(nbins) then nbins = max(bins)+1 h = histogram(bins,min=0,max=nbins-1,reverse=ri) if n_elements(h) ne nbins then message,'NBINS does not match h[]' endif nbins = n_elements(h) nan = fill_nan(a[0]) ;avg = replicate(nan,nbins) minimum = replicate(nan,nbins) avg = make_array(value=nan,dimen=size(h,/dimen)) std = avg whn0 = where(h ne 0,count) for j=0l,count-1 do begin i = whn0[j] ind = ri[ ri[i]: ri[i+1]-1 ] if n_elements(ind) ne h[i] then dprint ,'Histogram error' avg[i] = average(a[ind],stdev=s,nan=rnan) if arg_present(minimum) && size(/type,a) lt 7 then minimum[i] = min(a[ind]) std[i] = s endfor return,avg end