我有一套水質數據,我試圖確定是否有季節性趨勢。我已受季節分類的數據和我的數據是這樣的:固定效應分離使用nlme重複測量分析R
> Cond
Site Cond Season Watershed logCond
1 BICO201 41.86667 Spring BICO 1.621868
2 BICO301 53.16000 Spring BICO 1.725585
3 MIDO301 42.63333 Spring MIDO 1.629749
4 MIDO601 52.10000 Spring MIDO 1.716838
5 MIDO704 82.70000 Spring MIDO 1.917506
6 MIDO801 74.36667 Spring MIDO 1.871378
7 MIDO802 73.43333 Spring MIDO 1.865893
8 MIDO803 85.72000 Spring MIDO 1.933082
9 NORO401 43.30000 Spring NORO 1.636488
10 NORO502 132.05000 Spring NORO 2.120738
11 NORO503 61.36667 Spring NORO 1.787933
12 NORO517 142.40000 Spring NORO 2.153510
13 NORO520 95.20000 Spring NORO 1.978637
14 NORO527 81.08000 Spring NORO 1.908914
15 NORO601 479.75000 Spring NORO 2.681015
16 BICO201 47.73333 Summer BICO 1.678822
17 BICO301 58.46667 Summer BICO 1.766908
18 MIDO301 45.75000 Summer MIDO 1.660391
19 MIDO601 51.80000 Summer MIDO 1.714330
20 MIDO704 112.30000 Summer MIDO 2.050380
21 MIDO801 90.10000 Summer MIDO 1.954725
22 MIDO802 74.58000 Summer MIDO 1.872622
23 MIDO803 112.70000 Summer MIDO 2.051924
24 NORO401 71.40000 Summer NORO 1.853698
25 NORO502 192.88000 Summer NORO 2.285287
26 NORO503 80.42500 Summer NORO 1.905391
27 NORO517 156.50000 Summer NORO 2.194514
28 NORO520 114.22500 Summer NORO 2.057761
29 NORO527 109.00000 Summer NORO 2.037426
30 NORO601 420.00000 Summer NORO 2.623249
31 BICO201 46.85000 Fall BICO 1.670710
32 BICO301 55.43333 Fall BICO 1.743771
33 MIDO301 42.52500 Fall MIDO 1.628644
34 MIDO601 69.26667 Fall MIDO 1.840524
35 MIDO704 102.40000 Fall MIDO 2.010300
36 MIDO801 81.67500 Fall MIDO 1.912089
37 MIDO802 62.05000 Fall MIDO 1.792742
38 MIDO803 86.90000 Fall MIDO 1.939020
39 NORO401 62.85000 Fall NORO 1.798305
40 NORO502 149.60000 Fall NORO 2.174932
41 NORO503 57.90000 Fall NORO 1.762679
42 NORO517 92.90000 Fall NORO 1.968016
43 NORO520 118.31667 Fall NORO 2.073046
44 NORO527 123.15000 Fall NORO 2.090434
45 NORO601 522.33333 Fall NORO 2.717948
46 BICO201 101.96000 Winter BICO 2.008430
47 BICO301 69.47500 Winter BICO 1.841829
48 MIDO301 43.58333 Winter MIDO 1.639320
49 MIDO601 49.78000 Winter MIDO 1.697055
50 MIDO704 94.73333 Winter MIDO 1.976503
51 MIDO801 76.28000 Winter MIDO 1.882411
52 MIDO802 65.86667 Winter MIDO 1.818666
53 MIDO803 119.13333 Winter MIDO 2.076033
54 NORO401 54.20000 Winter NORO 1.733999
55 NORO502 171.76000 Winter NORO 2.234922
56 NORO503 83.76667 Winter NORO 1.923071
57 NORO517 191.07500 Winter NORO 2.281204
58 NORO520 118.31667 Winter NORO 2.073046
59 NORO527 123.15000 Winter NORO 2.090434
60 NORO601 576.00000 Winter NORO 2.760422
我試圖用隨季節作爲我的固定效果和網站作爲隨機效應的混合效果運行重複測量分析。我使用NLME包和我的代碼如下所示:
> mod.1.2<-lme(Cond~Season, random=~1|Site,data=Cond)
然後我跑我的模型的摘要,並得到如下的輸出:
> summary(mod.1.2)
Linear mixed-effects model fit by REML
Data: Cond
AIC BIC logLik
595.4271 607.5792 -291.7136
Random effects:
Formula: ~1 | Site
(Intercept) Residual
StdDev: 111.1618 22.68229
Fixed effects: Cond ~ Season
Value Std.Error DF t-value p-value
(Intercept) 111.61000 29.293255 42 3.810092 0.0004
SeasonSpring -8.86822 8.282401 42 -1.070731 0.2904
SeasonSummer 4.24733 8.282401 42 0.512814 0.6108
SeasonWinter 17.66200 8.282401 42 2.132474 0.0389
Correlation:
(Intr) SsnSpr SsnSmm
SeasonSpring -0.141
SeasonSummer -0.141 0.500
SeasonWinter -0.141 0.500 0.500
Standardized Within-Group Residuals:
Min Q1 Med Q3 Max
-3.3746755 -0.3431503 -0.0313137 0.3702357 2.9115215
Number of Observations: 60
Number of Groups: 15
我很困惑,因爲R的分手我的固定因素進入不同的季節,但我期待我的輸出只是給我一個價值/ StdDev/DF/p值的所有季節。
我想知道這是不是我誤解了lme的工作原理(我對R來說是非常新的),或者如果在我的公式中需要包含某些內容/適用於我的數據集以便在該級別完成分析所有季節。
我已經閱讀了一些關於解釋lme輸出的板子,但是我無法弄清楚我將如何解釋我目前得到的輸出,因爲季節是分開的。
我也試圖找到一個適當的事後測試。
任何幫助將不勝感激,謝謝!
非常感謝!實際上,我已將它的日誌轉換爲我的實際分析,爲簡單起見,我只是使用了我的初始公式! – Rose