> # Multinomial logit model using R
> mdata <-read.table("rate.csv", header=T, sep=",")
> y  <-mdata[,1]
> x1 <-mdata[,2]
> x2 <-mdata[,3]
> x3 <-mdata[,4]
> x4 <-mdata[,5]
> x5 <-mdata[,6]
> x6 <-mdata[,7]
> mydata <-data.frame(y,x1,x2,x3,x4,x5,x6)
> library(mlogit)
> mymldata <- mlogit.data(mydata, shape = "wide", choice ="y")
> mymlogit <-mlogit(y ~ 1 | x1 + x2 + x3 + x4 + x5 + x6, data=mymldata, reflevel="2")
> summary(mymlogit)

Call:
mlogit(formula = y ~ 1 | x1 + x2 + x3 + x4 + x5 + x6, data = mymldata, 
    reflevel = "2", method = "nr", print.level = 0)

Frequencies of alternatives:
      2       1       3 
0.33333 0.33333 0.33333 

nr method
8 iterations, 0h:0m:0s 
g'(-H)^-1g = 1.11E-07 
gradient close to zero 

Coefficients :
                 Estimate  Std. Error z-value  Pr(>|z|)    
1:(intercept) -37.5289770   8.8928380 -4.2201 2.442e-05 ***
3:(intercept)  48.6242495  12.9069413  3.7673 0.0001650 ***
1:x1            3.0236758   0.6873713  4.3989 1.088e-05 ***
3:x1           -4.1783254   1.0636135 -3.9284 8.550e-05 ***
1:x2            0.0719485   0.1953101  0.3684 0.7125893    
3:x2           -0.1579313   0.0490709 -3.2184 0.0012889 ** 
1:x3            0.1799104   0.0537495  3.3472 0.0008163 ***
3:x3           -0.2253704   0.0695072 -3.2424 0.0011853 ** 
1:x4            0.3470515   0.2607448  1.3310 0.1831888    
3:x4           -0.4081538   0.2914427 -1.4005 0.1613757    
1:x5           -0.0507558   0.0543179 -0.9344 0.3500871    
3:x5            0.0389759   0.0418980  0.9303 0.3522376    
1:x6           -0.0197212   0.0115403 -1.7089 0.0874712 .  
3:x6           -0.0028068   0.0119678 -0.2345 0.8145732    
---
Signif. codes:  0 �e***�f 0.001 �e**�f 0.01 �e*�f 0.05 �e.�f 0.1 �e �f 1

Log-Likelihood: -45.406
McFadden R^2:  0.65558 
Likelihood ratio test : chisq = 172.85 (p.value = < 2.22e-16)