Checking validity for the model “Ozone_lm1” by way of standardised residuals, leverages, and Cook's distances

Prepared by BrailleR

Basic summary measures

Counts

116 values in all, made up of
114 unique values,
116 observed, and
0 missing values.

Measures of location

Mean Value
All data 0.0060427
Trimmed 5% -0.0152765
Trimmed 10% -0.0346718

Quantiles

Quantile Value
0% Minimum -1.9881298
25% Lower Quartile -0.7150325
50% Median -0.1840261
75% Upper Quartile 0.5804802
100% Maximum 3.6474063

Measures of spread

Measure Value
IQR 1.2955127
Standard deviation 1.0175268
Variance 1.0353609

Basic univariate graphs

Histogram

The histogram

This is a histogram, with the title: Histogram of Residuals 
 "Residuals" is marked on the x-axis.
Tick marks for the x-axis are at: -2, and 4 
There are a total of 116 elements for this variable.
Tick marks for the y-axis are at: 0, 10, 20, 30, and 40 
It has 6 bins with equal widths, starting at -2 and ending at 4 .
The mids and counts for the bins are:
mid = -1.5  count = 19 
mid = -0.5  count = 47 
mid = 0.5  count = 31 
mid = 1.5  count = 14 
mid = 2.5  count = 4 
mid = 3.5  count = 1

Boxplot

The boxplot

This graph has a boxplot printed horizontally
with the title: Boxplot of Residuals
No label appears on the x-axis.
No label appears on the y-axis.
Tick marks for the x-axis are at: -2, and 3 
This variable has 116 values.
An outlier is marked at: 3.647406 
The whiskers extend to -1.98813 and 2.384745 from the ends of the box, 
which are at -0.7168757 and 0.6329842 
The median, -0.1840261 is 39 % from the left end of the box to the right end.
The right whisker is 1.38 times the length of the left whisker.

Assessing normality

Formal tests for normality

Statistic P Value
Shapiro-Wilk 0.9642 0.0034
Anderson-Darling 1.0770 0.0076
Cramer-von Mises 0.1778 0.0101
Lilliefors (Kolmogorov-Smirnov) 0.0886 0.0259
Pearson chi-square 11.2069 0.4261
Shapiro-Francia 0.9633 0.0040

Normality plot

The normality plot

Formal tests of moments

Statistic Z P Value
D'Agostino skewness 0.7397 3.132 0.0017
Anscombe-Glynn kurtosis 3.5490 1.359 0.1740

Regression diagnostic plots

Standardised residuals

Standardised residuals plotted against fitted values

    1  2  3  4 Sum
4   2  6 10 11  29
3   3  9  6  3  21
2   0 15 16  5  36
1   0  3 26  1  30
Sum 5 33 58 20 116

Standardised residuals plotted against order

     1  2  3  4 Sum
4    4 10 11  4  29
3    6  6  6  3  21
2    9  7  8 12  36
1   10  6  4 10  30
Sum 29 29 29 29 116

standardised residuals plotted against lagged residuals

     1  2  3  4 Sum
4    5  7  4 13  29
3    5  7  4  5  21
2   12 11  7  5  35
1    8 10  6  6  30
Sum 30 35 21 29 115

The lag 1 autocorrelation of the standardised residuals is 0.1899274.

Influence

Standardised residuals plotted against leverages

      1  2 3 4 Sum
4    22  5 0 2  29
3    16  4 1 0  21
2    35  1 0 0  36
1    30  0 0 0  30
Sum 103 10 1 2 116

8 points have excessive leverage.
0 points have Cook's distances greater than one.

Outliers and influential observations

Ozone Wind Fit St.residual Leverage Cooks.distance
9 8 20.1 -14.700655 0.8933795 0.0799404 0.0347346
18 6 18.4 -5.264086 0.4369840 0.0582217 0.0059447
22 11 16.6 4.727575 0.2408120 0.0395121 0.0012027
48 37 20.7 -18.031209 2.2148620 0.0885448 0.2303889
62 135 4.1 74.114111 2.3847448 0.0312120 0.0879931
86 108 8.0 52.465512 2.1428244 0.0109800 0.0247097
101 110 8.0 52.465512 2.2233100 0.0109800 0.0265215
117 168 3.4 77.999757 3.6474063 0.0370344 0.2308992
121 118 2.3 84.105772 1.3163831 0.0475312 0.0429615
126 73 2.8 81.330310 -0.3203899 0.0425558 0.0022993
148 14 16.6 4.727575 0.3560969 0.0395121 0.0026284