生态方向做模型，R 提醒更改变量，求指导

Planthopper

2197

收藏 2017-06-16

首先感谢一下
全数据见附件
部分如下

id	year	distance1	sum1	sum
1	2006	183.21	5148	3.711722892
2	2006	442	1360	3.133858125
3	2006	187.33	11455	4.059033005
4	2006	795.74	287	2.459392488
5	2006	684.69	1485	3.172018809
6	2006	552.89	2382	3.377124042
7	2006	813.09	553	2.743509765
8	2006	525.11	53149	4.725503269
9	2006	510.27	29680	4.472478529
10	2006	806.29	201	2.305351369
11	2006	728.41	207	2.318063335
12	2006	770.91	42170	4.6250139
13	2006	766.86	11507	4.060999853
14	2006	987.29	563	2.751279104
15	2006	1250.55	703	2.847572659
16	2006	1369.28	6	0.84509804
17	2006	1246.47	4551	3.658202253
18	2006	1157.04	2032	3.308137379
19	2006	1178.97	5850	3.767230098
20	2006	1435.69	49	1.698970004
21	2006	1264.55	1392	3.143951116
22	2006	1358.17	172	2.238046103
23	2006	1579.72	216	2.336459734
24	2006	1561.39	119	2.079181246
25	2006	1509.79	0	0
26	2006	1563.55	52	1.72427587
27	2006	1456.39	57	1.763427994
28	2006	1472.95	161	2.209515015
29	2006	1398.46	1920	3.283527365
30	2006	1632.7	8	0.954242509
31	2006	1590.86	3	0.602059991
32	2006	902.86	705	2.848804701
33	2006	902.35	90	1.959041392
34	2007	442	38	1.591064607
35	2007	187.33	4061	3.608739919
36	2007	183.21	5148	3.711722892
37	2007	525.11	3067	3.486855355
38	2007	795.74	118	2.075546961
39	2007	684.69	741	2.870403905
40	2007	576.22	56	1.755874856
41	2007	813.09	420	2.624282096
42	2007	510.27	67498	4.829297339
43	2007	551.74	921	2.964730921
44	2007	552.89	0	0
45	2007	806.29	1086	3.036229544
46	2007	770.91	104824	5.020464871
47	2007	987.29	6	0.84509804
48	2007	0.99	6892	3.838408278
49	2007	1435.69	0	0
50	2007	1369.28	792	2.899273187
51	2007	1246.47	4009	3.603144373
52	2007	1250.55	8568	3.932930143
53	2007	1157.04	327	2.515873844
54	2007	1178.97	542	2.73479983
55	2007	1358.17	340	2.532754379
56	2007	1579.72	102	2.012837225
57	2007	1561.39	114	2.06069784
58	2007	1509.79	619	2.792391689
59	2007	1632.7	735	2.866877814
60	2007	1590.86	74	1.875061263
61	2007	1442.58	357	2.553883027
62	2007	1251.58	98	1.995635195
63	2007	1456.39	36	1.568201724
64	2007	1472.95	120	2.08278537
65	2007	902.86	48	1.69019608

采用GLMM模型，想看sum与distance的关系，sum是取对数后的sum1
在R里面用的是M1 <- glmer(sum1 ~ distance1 + (1|year),  family = poisson)
运行之后出现的问题如下
Warning messages:
1: In checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv,  :
  Model failed to converge with max|grad| = 0.236049 (tol = 0.001, component 1)
2: In checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv,  :
  Model is nearly unidentifiable: very large eigenvalue
- Rescale variables?;Model is nearly unidentifiable: large eigenvalue ratio
- Rescale variables?

应该是需要调整变量，但是不太清楚该如何调整，求大牛知道，跪谢跪谢~~~~