R语言Bass模型进行销售预测
原文链接:http://tecdat.cn/?p=3937
BASS扩散模型
BASS扩散模型三个参数:最终购买产品的总人数m; 创新系数p; 和模仿系数q,
nls(Sales~M*(((P+Q)^2/P)\*exp(-(P+Q)\*T79))/(1+(Q/P)\*exp(-(P+Q)\*T79))^2,
## Parameters:
## Estimate Std. Error t value Pr(>|t|)
## M 6.80e+04 3.13e+03 21.74 1.1e-07 ***
## P 6.59e-03 1.43e-03 4.61 0.0025 **
## Q 6.38e-01 4.14e-02 15.41 1.2e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 727 on 7 degrees of freedom
##
## Number of iterations to convergence: 8
## Achieved convergence tolerance: 7.32e-06
# 获得系数
Bcoef <- coef(Bass.nls)
m <- Bcoef\[1\]
p <- Bcoef\[2\]
q <- Bcoef\[3\]
将M的起始值设置为记录的总销售额。
ngete <- exp(-(p + q) * Tdelt)
# 绘制概率密度函数
Bpdf <- m * ((p + q)^2/p) * ngete/(1 + (q/p) * ngete)^2
plot(Tdelt, Bpdf
plot(Tdelt, Bcdf, xlab = "Year from 1979"
#当q = 0时,只有创新者没有模仿者。
Ipdf <- m * ((p + 0)^2/p) * exp(-(p + 0) * Tdelt)/(1 + (0/p) * exp(-(p + 0) *
col = "red")lines(Tdelt, Impdf, col = "green")lines(Tdelt, Ipdf, col = "blue")
#当q = 0时
Icdf <- m * (1 - exp(-(p + 0) * Tdelt))/(1 + (0/p) * exp(-(p + 0) * Tdelt))
lines(Tdelt, Icdf, col = "blue")