我偶然发现了很多类似问题，但找不到任何答案。我的问题是在这篇文章末尾复制的数据，它再现了我试图使用plm包通过随机效应模型获得的数据集。
错误如下：
# data provided at end of post
library(plm)
plm(y ~ x + factor(year), index = "panel", model = "random", data = df)
# Error in solve.default(crossprod(ZBeta)) :
# Lapack routine dgesv: system is exactly singular: U[15,15] = 0
问题在于factor(year)部分。我不明白的是，相同的模型和数据在Stata:中完美工作
xtreg y x i.year, re sa
// output omitted
(上面的sa选项选择了plm默认使用的Swamy Arora估计技术。感谢Nir Graham的回答，这让我观察到了这一点。(
删除我在这个(非常不平衡的(小组中只有一个观察结果的年份并没有帮助。如有任何帮助，我们将不胜感激！
> dput(df)
structure(list(y = c(0.32, 0.51, 0.26, 0.99, 0.59, 0.43, 0.6,
0.86, 1, 0.97, 0.89, 0.63, 0.55, 0.58, 0.26, 0.69, 0.87, 0.17,
0.09, 0, 0.87, 0.39, 0.36, 0.73, 0.13, 0.61, 0.36, 0.64, 0.72,
0.95, 0.8, 0.96, 0.32, 0.91, 0.77, 0.14, 0.37, 0.57, 0.81, 0.98,
0.5, 0.23, 0.8, 0.04, 0.84, 0.12, 0.56, 0.22, 0.49, 0.65, 0.59,
0.98, 0.71, 0.58, 0.75, 0.77, 0.49, 0.72, 0.29, 0.2, 0.67, 0.06,
0.36, 0.44, 0.65, 0.29, 0.85, 0.75, 0.2, 0.44, 0.7, 0.54, 0.19,
0.47, 0.83, 0.47, 0.23, 0.43, 0.6, 0.48, 0.63, 0.95, 1, 0.46,
0.28, 0.88, 0.82, 0.71, 0.57, 0.25, 0.78, 0.07, 0.45, 0.7, 0.08,
0.2, 0.5, 0.13, 0.56, 0.12, 0.08, 0.29, 0.89, 0.37, 0.96, 0.83,
0.81, 0.02, 0.96, 0.83, 0.51, 0.04, 0.04, 0.06, 0.44, 0.61, 0.99,
0.83, 0.31, 0.82, 0.12, 0.18, 0.89, 0.23, 0.46, 0.73, 0.76, 0.49,
0.32, 0.87, 0.11, 0.01, 0.96, 0.86, 0.91, 0.68, 0.8, 0.63, 0.94,
1, 0.59, 0.5, 0.01, 0.48, 0.86, 0.92, 0.07, 0.15, 0.07, 0.33,
0.6, 0.52, 0.12, 0.59, 0.56, 0.56, 0.55, 0.18, 0.11, 0.16, 0.27,
0.06, 0.62, 0.34, 0.69, 0.87, 0.32, 0.31, 0.1, 0.44, 0.99, 0.96,
0.72, 0.19, 0.81), x = c(0.25, 0.41, 0.55, 0.77, 0.95, 0.2, 0.36,
0.58, 0.27, 0.56, 0.53, 0.88, 0.55, 0.43, 0.19, 0.54, 0.2, 0.37,
0.18, 0.09, 0.26, 0.15, 0.75, 0.08, 0.55, 0.06, 0.23, 0.9, 0.12,
0.51, 0.58, 0.54, 0.88, 0.24, 0.9, 0.85, 0.32, 0.43, 0.66, 0.12,
0.09, 0.75, 0.5, 0.11, 0.07, 0.04, 0.6, 0.96, 0.39, 0.61, 0.23,
0.28, 0.45, 0.55, 0.52, 0.99, 0.96, 0.64, 0.31, 0.47, 0.01, 0.56,
0.7, 0.88, 0.13, 0.87, 0.2, 0.62, 0.42, 0.85, 0.5, 0.22, 0.52,
0.15, 0.31, 0.23, 0.09, 0.76, 0.56, 0.29, 0.42, 0.87, 0.75, 0.78,
0.67, 0.94, 0.69, 0.74, 0.07, 0.22, 0.47, 0.52, 0.85, 0.28, 0.47,
0.39, 0.34, 0.94, 0.14, 0.5, 0.16, 0.2, 0.22, 0.71, 0.66, 0.68,
0.54, 0.24, 0.04, 0.1, 0.44, 0.54, 0.23, 0.53, 0.24, 0.14, 0.99,
0.18, 0.93, 0.99, 0.49, 0.39, 0.78, 0.41, 0.31, 0.11, 0.75, 0.59,
0.85, 0.31, 0.8, 0.21, 0.67, 0.31, 0.21, 0.88, 0.84, 0.32, 0.36,
0.89, 0.4, 0.82, 0.54, 0.18, 0.4, 0.71, 0.28, 0.83, 0.78, 0.07,
0.93, 0.47, 0.44, 0.49, 0.71, 0.69, 0, 0.47, 0.72, 0.06, 0.13,
0.65, 0.12, 0.26, 0.67, 0.8, 0.4, 0.82, 0.22, 0.16, 0.32, 0.01,
0.53, 0.26, 0.99), panel = c("p1", "p1", "p1", "p1", "p1", "p1",
"p2", "p2", "p2", "p2", "p2", "p2", "p2", "p2", "p2", "p3", "p3",
"p3", "p3", "p4", "p4", "p4", "p4", "p5", "p5", "p5", "p5", "p6",
"p6", "p6", "p6", "p6", "p6", "p6", "p6", "p6", "p7", "p7", "p7",
"p7", "p7", "p7", "p7", "p7", "p7", "p8", "p8", "p8", "p8", "p8",
"p8", "p8", "p8", "p9", "p9", "p9", "p9", "p10", "p10", "p10",
"p10", "p10", "p10", "p11", "p11", "p11", "p11", "p11", "p11",
"p11", "p11", "p11", "p12", "p12", "p12", "p12", "p12", "p12",
"p12", "p12", "p12", "p13", "p13", "p13", "p13", "p13", "p13",
"p13", "p13", "p13", "p14", "p14", "p14", "p15", "p15", "p15",
"p15", "p16", "p16", "p16", "p16", "p16", "p16", "p16", "p16",
"p16", "p17", "p17", "p17", "p17", "p17", "p17", "p17", "p17",
"p17", "p18", "p18", "p18", "p18", "p19", "p19", "p19", "p19",
"p19", "p19", "p19", "p19", "p19", "p20", "p20", "p20", "p20",
"p21", "p21", "p21", "p21", "p22", "p22", "p22", "p22", "p22",
"p22", "p22", "p22", "p22", "p23", "p23", "p23", "p23", "p24",
"p24", "p24", "p24", "p24", "p24", "p24", "p24", "p25", "p25",
"p25", "p25", "p26", "p26", "p26", "p26", "p27", "p27", "p27",
"p27", "p28", "p28", "p28", "p28", "p28", "p28"), year = c(8,
9, 10, 12, 14, 15, 1, 3, 5, 6, 9, 10, 12, 14, 15, 11, 12, 14,
15, 10, 12, 14, 15, 10, 12, 14, 15, 1, 3, 5, 6, 9, 10, 12, 14,
15, 1, 3, 5, 6, 9, 10, 12, 14, 15, 4, 5, 6, 9, 10, 12, 14, 15,
10, 12, 14, 15, 8, 9, 10, 12, 14, 15, 1, 3, 5, 6, 9, 10, 12,
14, 15, 1, 3, 5, 6, 9, 10, 12, 14, 15, 2, 3, 5, 6, 9, 10, 12,
14, 15, 13, 14, 15, 10, 12, 14, 15, 1, 3, 5, 6, 9, 10, 12, 14,
15, 1, 3, 5, 6, 9, 10, 12, 14, 15, 10, 12, 14, 15, 1, 3, 5, 6,
9, 10, 12, 14, 15, 10, 12, 14, 15, 10, 12, 14, 15, 1, 3, 5, 6,
9, 10, 12, 14, 15, 10, 12, 14, 15, 4, 5, 6, 9, 10, 12, 14, 15,
11, 12, 14, 15, 10, 12, 14, 15, 10, 12, 14, 15, 7, 9, 10, 12,
14, 15)), row.names = c(NA, -175L), class = "data.frame")
p.S.克罗斯贴出了包裹的问题。
plm有一些算法；也许random.method="；nerlove；让你更接近你想要的