Question

I have a directed dyadic data frame that looks like this.

       ccode.a ccode.b year  mindist int.a2b
167603     220     570 1976 8.985324       0
624316     781     770 1976 7.914206       0
593896     740     620 1976 9.128673       0
669443     900     660 1976 9.235735       1
323404     434      20 1976 8.764419       0
353101     451     432 1976 5.390160       0
37109       53     700 1976 9.373724       0
624853     790       2 1976 8.949047       0
242472     355      53 1976 9.030233       0
35129       53     350 1976 9.006963       0
481129     600     140 1976 8.377329       0
621802     781     310 1976 8.885368       0
236655     350     600 1976 7.587584       0
192503     290     950 1976 9.636472       1
464722     580      94 1976 9.551535       0
550661     694      42 1976 9.377167       0
585022     712     625 1976 8.586424       0
637007     812       2 1976 8.902645       0
539818     678     572 1976 8.402542       0
690214     950     900 1976 7.859377       1
135804     160     135 1976 6.314121       0
554291     694     811 1976 8.603649       0
496453     620     680 1976 7.565657       0
68160       90     220 1976 9.003781       0
605932     770     482 1976 8.355514       0
509185     640     660 1976 4.873909       0
24928       42     830 1976 9.774045       0
454705     570     666 1976 8.740049       0
92800      100      51 1976 6.631003       1
492025     616     651 1976 7.140152       0
598335     750     663 1976 7.997375       0
485329     615     130 1976 8.951390       0
330093     435     520 1976 8.560956       0
74135       91     570 1976 9.453192       0
465351     580     235 1976 8.989404       1
227129     345     165 1976 9.296795       0
488046     615     696 1976 8.298939       0
381548     483     375 1976 8.325534       0
237750     350     840 1976 9.103709       0
402138     510     265 1976 8.688151       0
372394     475     950 1976 9.798338       0
216445     338     570 1976 8.890926       1
598308     750     660 1976 8.082419       0
613237     775     235 1976 9.100852       0
310790     420      42 1976 8.617836       0
611272     771     696 1976 8.065013       0
183450     235     551 1976 8.739090       0
659939     840     590 1976 8.879307       0
506308     640      70 1976 9.232063       0
188756     290      91 1976 9.120557       1

I want to create a matrix of all of the pairs of ccode.a and ccode.b,

d.sender <- matrix(0, nrow = 192, ncol = 192, dimnames = list(unique(dyad$ccode.a), unique(dyad$ccode.a)))

And then assign the value of int.a2b (I’ll have to do this for a bunch of variables) to the appropriate cell in the matrix. This will be done separately for each year in the data set. Since the data frame is directed dyadic I’ll also have a receiver matrix of the same dimension that that mirrors the sender matrix I’m trying to create now.

Answer 1

If I am correct in understanding what you want, here is one way:

 x <- data.frame(a=(1:3),b=(4:6),val=c(8,3,7))
 library(reshape2)

 acast(x, a~b, value.var="val")
#   4  5  6
#1  8 NA NA
#2 NA  3 NA
#3 NA NA  7