Good morning, I'm dealing with graphic representation of missing data imputeted via mean imputation method. This is the dataset I'm working on:
> data2
age fev ht sex smoke
1 9 1.708 145 1 1
2 8 1.724 171 1 1
3 7 1.720 138 1 1
4 9 1.558 135 2 1
5 9 1.895 145 2 1
6 8 2.336 155 1 1
7 6 1.919 147 1 1
8 6 1.415 142 1 1
9 8 1.987 149 1 1
10 9 1.942 152 1 1
11 6 1.602 135 1 1
12 8 1.735 137 2 1
13 8 2.193 149 1 1
14 8 2.118 154 2 1
15 8 2.258 147 2 1
16 7 1.932 135 2 1
17 5 1.472 127 2 1
18 6 1.878 NA 1 1
19 9 2.352 150 2 1
20 9 2.604 156 2 1
21 5 1.400 124 1 1
22 5 1.256 133 1 1
23 4 0.839 122 1 1
24 7 2.578 159 2 1
25 9 2.988 165 1 1
26 3 1.404 131 2 1
27 9 2.348 152 2 1
28 5 1.755 132 2 1
29 8 2.980 152 1 1
30 9 2.100 152 1 1
31 5 1.282 124 1 1
32 9 3.000 166 2 1
33 8 2.673 152 1 1
34 7 2.093 146 1 1
35 5 1.612 132 1 1
36 8 2.175 150 1 1
37 9 2.725 150 2 1
38 8 2.071 140 2 1
39 8 1.547 145 2 1
40 8 2.004 145 2 1
41 9 3.135 152 1 1
42 8 2.420 150 2 1
43 5 1.776 130 2 1
44 8 1.931 145 1 1
45 5 1.343 127 1 1
46 9 2.076 145 1 1
47 7 1.624 137 2 1
48 8 1.344 133 1 1
49 6 1.650 140 2 1
50 8 2.732 154 2 1
51 5 2.017 138 2 1
52 9 2.797 156 1 1
53 9 NA 157 2 1
54 8 1.703 138 2 1
55 6 1.634 137 2 1
56 9 2.570 145 2 1
57 9 3.016 159 1 1
58 7 2.419 152 1 1
59 4 1.569 127 1 1
60 8 1.698 146 1 1
61 8 2.123 152 2 1
62 8 2.481 152 1 1
63 6 1.481 130 1 1
64 4 1.577 124 1 1
65 8 1.940 150 2 1
66 6 1.747 146 2 1
67 9 2.069 147 2 1
68 7 1.631 141 1 1
69 5 1.536 132 1 1
70 9 2.560 154 1 1
71 8 1.962 145 2 1
72 8 2.531 147 1 1
73 9 2.715 152 2 1
74 9 2.457 150 2 1
75 9 2.090 151 2 1
76 7 1.789 142 2 1
77 5 1.858 135 2 1
78 5 1.452 130 2 1
79 9 NA 175 2 1
80 6 1.719 135 1 1
81 7 2.111 145 1 1
82 6 1.695 135 1 1
83 8 2.211 160 2 1
84 8 1.794 138 2 1
85 7 1.917 147 1 1
86 8 2.144 NA 1 1
87 7 1.253 132 2 1
88 9 2.659 156 2 1
89 5 1.580 133 2 1
90 9 2.126 157 2 1
91 9 3.029 156 1 1
92 9 2.964 164 2 1
93 7 1.611 NA 2 1
94 8 2.215 152 1 1
95 8 2.388 152 1 1
96 9 2.196 155 2 1
97 9 1.751 147 2 1
98 9 2.165 156 2 1
99 7 1.682 140 2 1
100 8 1.523 140 2 1
101 8 1.292 132 1 1
102 7 1.649 137 2 1
103 9 2.588 160 2 1
104 4 0.796 119 2 1
105 9 2.574 154 1 1
106 6 1.979 142 2 1
107 8 2.354 149 2 1
108 6 1.718 140 2 1
109 7 1.742 149 1 1
110 7 1.603 130 1 1
111 8 2.639 151 1 1
112 7 1.829 137 1 1
113 7 2.084 147 2 1
114 7 2.220 147 2 1
115 7 1.473 133 1 1
116 8 2.341 154 1 1
117 7 1.698 138 1 1
118 5 1.196 118 1 1
119 8 1.872 144 1 1
120 7 2.219 140 2 1
121 9 2.420 145 2 1
122 7 1.827 138 1 1
123 7 1.461 137 1 1
124 6 1.338 NA 2 1
125 8 2.090 145 2 1
126 8 1.697 150 1 1
127 8 1.562 140 2 1
128 9 2.040 141 1 1
129 7 1.609 131 1 1
130 8 2.458 155 1 1
131 9 2.650 161 2 1
132 8 1.429 146 2 1
133 8 1.675 135 2 1
134 9 1.947 144 1 1
135 8 2.069 137 2 1
136 6 1.572 132 2 1
137 6 1.348 135 2 1
138 8 2.288 156 1 1
139 9 1.773 149 2 1
140 5 0.791 132 1 1
141 7 1.905 147 2 1
142 9 2.463 155 1 1
143 6 1.431 130 2 1
144 9 2.631 157 1 1
145 9 3.114 164 2 1
146 9 2.135 149 2 1
147 6 1.527 133 2 1
148 8 2.293 147 1 1
149 9 3.042 168 1 1
150 8 2.927 161 2 1
151 8 2.665 163 1 1
152 9 2.301 149 2 1
153 9 2.460 163 2 1
154 9 2.592 154 1 1
155 7 1.750 140 1 1
156 8 1.759 135 2 1
157 6 1.536 122 2 1
158 9 2.259 149 1 1
159 9 2.048 164 1 1
160 9 2.571 154 2 1
161 7 2.046 142 2 1
162 8 1.780 149 1 1
163 5 1.552 137 1 1
164 8 1.953 147 1 1
165 9 2.893 164 2 1
166 6 1.713 128 2 1
167 9 2.851 152 1 1
168 6 1.624 131 2 1
169 8 2.631 150 2 1
170 5 1.819 135 2 1
171 7 1.658 135 2 1
172 7 2.158 136 2 1
173 4 1.789 132 2 1
174 9 3.004 163 1 1
175 8 2.503 160 2 1
176 9 1.933 147 1 1
177 9 2.091 149 1 1
178 9 2.316 NA 1 1
179 5 1.704 NA 1 1
180 9 1.606 146 1 1
181 7 1.165 119 2 1
182 6 2.102 141 1 1
183 9 2.320 145 1 1
184 9 2.230 155 2 1
185 9 1.716 141 2 1
186 7 1.790 136 2 1
187 5 1.146 127 1 1
188 8 2.187 156 1 1
189 9 2.717 156 2 1
190 7 1.796 140 2 1
191 9 1.953 147 2 2
192 8 1.335 144 1 1
193 9 2.119 145 2 1
194 6 1.666 132 2 1
195 6 1.826 133 2 1
196 8 2.709 159 1 1
197 9 2.871 165 2 1
198 5 1.092 127 1 1
199 6 2.262 146 2 1
200 6 2.104 144 2 1
I've used the following code to return back the observed vs the imputed data scenario and the sided scatterplot of Y="fev" versus X="age".
1. FIRST GRAPH
library(lattice)
par(mfrow=c(1,2))
breaks <- seq(-20, 200, 10)
nudge <- 1
lwd <- 1.5
x <- matrix(c(breaks-nudge, breaks+nudge), ncol=2, nrow = 46)
obs <- data2[,"fev"]
mis <- imp$imp$fev[,1]
fobs <- c(hist(obs, breaks, plot=FALSE)$fev, 0)
fmis <- c(hist(mis, breaks, plot=FALSE)$fev, 0)
y <- matrix(c(fobs, fmis), ncol=2, nrow = 46)
matplot(x, y, type="s",
col=c(mdc(4),mdc(5)), lwd=2, lty=1,
xlim = c(0, 150), ylim = c(0,40), yaxs = "i",
xlab="fev",
ylab="Frequency")
box()
2. SECOND GRAPH
tp <- xyplot(imp, fev ~ age, na.groups=ici(imp),
ylab="fev", xlab="age",
cex = 0.75, lex=lwd, pch=19,
ylim = c(-20, 180), xlim = c(0,350))
print(tp, newpage = FALSE, position = c(0.48,0.08,1,0.92))
Although the code works well, I'm not so as to its validity, as I am supposed to have back a graphic results like those I let attached here enter image description here, whereas I'm keeping on getting a sort of graphs like these enter image description here
What do you think about? Any clue as to making the right cade out? Thanks for helping
