I have some data that I've fit a glm from the family quasibinomial. The model shows high significance for a number of different factors. However, I was wondering how can I get effect sizes for these different factors. Does anyone have any idea?
The model results:
Call:
glm(formula = freq ~ K + res_dist + K:trail_cost + trail_cost:res_dist:K,
family = "quasibinomial", data = data)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.2072 -0.3505 -0.1406 0.2714 1.7746
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.222e+00 1.786e-01 6.842 7.24e-11 ***
K -1.741e-05 2.107e-06 -8.265 1.19e-14 ***
res_distrandom -1.419e+00 2.386e-01 -5.949 1.02e-08 ***
K:trail_cost 1.381e-01 3.930e-02 3.515 0.000531 ***
K:res_distrandom:trail_cost 1.564e-01 2.532e-02 6.176 3.03e-09 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for quasibinomial family taken to be 0.2932451)
Null deviance: 126.028 on 230 degrees of freedom
Residual deviance: 66.616 on 226 degrees of freedom
AIC: NA
Number of Fisher Scoring iterations: 5
Related
I've run an Interrupted Time Series Analysis using a GLM and need to be able to exponentiate outcomes in order to validate. I have been recommended the emmeans package, but I'm not quite sure how to do it.
Base R summary is below:
summary(fit1a)
Call:
glm(formula = `Subject Total` ~ Quarter + int2 + time_since_intervention2,
family = "poisson", data = df)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.4769 -0.5111 0.1240 0.6103 0.9128
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 3.54584 0.09396 37.737 <0.0000000000000002 ***
Quarter -0.02348 0.01018 -2.306 0.0211 *
int2 -0.23652 0.21356 -1.108 0.2681
time_since_intervention2 -0.02624 0.04112 -0.638 0.5234
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for poisson family taken to be 1)
Null deviance: 63.602 on 23 degrees of freedom
Residual deviance: 13.368 on 20 degrees of freedom
AIC: 140.54
Number of Fisher Scoring iterations: 4
Cant really understand how to get started with emmeans. How would I code in order to get the estimate for Quarter, int2 and time_since_intervention2 on the response scale?
I am trying to determine if the hospital I work at should open a new unit for admissions. I intend to do this by correlating patient assigned unit and length of stay in days.
So far, I have determined that the continuous dependent variable (length of stay in days) should be log transformed, as well as the continuous age variable. Unit is categorical and contains 7 categories (unit 65, unit 66, unit 67, unit 75, unit 76, unit 77, unit 94). Gender is standard 0=male and 1=female.
My regression equation is as follows:
log_glm <- glm(log(days)~ gender + log(age) + unit, family=gaussian, data=data2022)
Model Summary:
glm(formula = log(days) ~ gender + log(age) + unit, family = gaussian,
data = data2022)
Deviance Residuals:
Min 1Q Median 3Q Max
-4.3606 -0.6712 0.0011 0.7146 2.8338
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.7416 0.9994 0.742 0.459064
gender 0.3257 0.2082 1.564 0.119566
log(age) 1.3809 0.2353 5.870 2.17e-08 ***
unitunit_66 0.1329 0.2994 0.444 0.657717
unitunit_67 1.2518 0.3334 3.755 0.000237 ***
unitunit_75 -1.8108 0.3067 -5.903 1.84e-08 ***
unitunit_76 0.2797 0.3120 0.896 0.371355
unitunit_77 -0.3540 0.3178 -1.114 0.266803
unitunit_94 0.1876 0.5350 0.351 0.726339
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for gaussian family taken to be 1.145772)
Null deviance: 429.95 on 181 degrees of freedom
Residual deviance: 198.22 on 173 degrees of freedom
(14 observations deleted due to missingness)
AIC: 552.03
Number of Fisher Scoring iterations: 2
All units at this hospital house patients with different needs. Is it appropriate to "compare" each unit in the summary output to the reference unit when all units are unique? Is this interpretation correct:
Unit 67 is statistically significant at the 0.05 level. The coefficient is 1.2518, so being assigned to unit 67 is associated with a length of stay in days 249.6631% ((exp(1.2518)-1100) higher on average than those assigned to unit 65.*
Doubting myself because that result seemed too high and interpretation incorrect, I tried a model with dummy variables for unit:
log_glm_age <- glm(log(days)~ gender + log(age) + unit_65 + unit_66 + unit_67 + unit_75 + unit_76 + unit_77 + unit_94, family=gaussian, data=data2022)
With output:
glm(formula = log(days) ~ gender + log(age) + unit_65 + unit_66 +
unit_67 + unit_75 + unit_76 + unit_77 + unit_94, family = gaussian,
data = data2022)
Deviance Residuals:
Min 1Q Median 3Q Max
-4.3606 -0.6712 0.0011 0.7146 2.8338
Coefficients: (1 not defined because of singularities)
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.92919 1.10243 0.843 0.400475
gender 0.32568 0.20819 1.564 0.119566
log(age) 1.38094 0.23525 5.870 2.17e-08 ***
unit_65 -0.18755 0.53500 -0.351 0.726339
unit_66 -0.05465 0.52964 -0.103 0.917936
unit_67 1.06427 0.55764 1.909 0.057979 .
unit_75 -1.99832 0.53701 -3.721 0.000268 ***
unit_76 0.09213 0.54694 0.168 0.866435
unit_77 -0.54160 0.52780 -1.026 0.306252
unit_94 NA NA NA NA
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for gaussian family taken to be 1.145772)
Null deviance: 429.95 on 181 degrees of freedom
Residual deviance: 198.22 on 173 degrees of freedom
(14 observations deleted due to missingness)
AIC: 552.03
Number of Fisher Scoring iterations: 2
But I am having doubts interpreting this. Since each patient is only admitted to one unit, is it appropriate to have all units in the model?
I don't know if it's because I am the only analyst in the entire hospital and can't discuss this with anyone who understands or if I'm having a weird day, but I can't figure out what is correct right now and could use some guidance.
I am trying to plot best fitting poisson distribution over a histogram to show over-dispersion in data. I came across a piece of code: The first part creates an histogram and calculates a Poisson model. So far so good.
hist(patents$ncit, nclas=14,col="light blue",prob=T,
xlab="Number of citations",ylab="",main="",
cex.lab=1.5,cex.axis=1.3)
glm(formula = ncit ~ 1, family = poisson, data = patents)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.7513 -1.7513 -0.4604 0.3596 6.4405
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.42761 0.01164 36.72 <2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for poisson family taken to be 1)
Null deviance: 13359 on 4808 degrees of freedom
Residual deviance: 13359 on 4808 degrees of freedom
AIC: 20350
Number of Fisher Scoring iterations: 6
In the second part of the code a best fitting poisson line is constructed. I do not understand where does the exp(0.32723)) comes from ?
lines(0:14,dpois(0:14,exp(0.32723)),col="red",lwd=2)
This is my glm model which I was able to create but I now want to do the Hosmer Lemeshow GOF test but get this error which I don't understand-
Call:
glm(formula = BC.result ~ Diabetic + Low.diastolic + Pulse, family =
"binomial", data = Data2)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.0805 -0.4559 -0.3144 -0.2437 2.8259
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -5.86311 0.98715 -5.939 2.86e-09 ***
Diabetic 1.21963 0.37395 3.262 0.001108 **
Low.diastolic 1.27095 0.35074 3.624 0.000291 ***
Pulse 0.02361 0.00780 3.027 0.002470 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 276.40 on 485 degrees of freedom
Residual deviance: 249.71 on 482 degrees of freedom
(14 observations deleted due to missingness)
AIC: 257.71
> hl<- hoslem.test(model3$BC.result, fitted(model3, g=10))
Error in model.frame.default(formula = cbind(y0 = 1 - y, y1 = y) ~ cutyhat)
:
variable lengths differ (found for 'cutyhat')
Can anyone explain this error with the Hosmer Lemeshow goodness of fit test?
Thanks
I have run a logistic regression, the summary of which I name. "score" Accordingly, summary(score) gives me the following
Deviance Residuals:
Min 1Q Median 3Q Max
-1.3616 -0.9806 -0.7876 1.2563 1.9246
Estimate Std. Error z value Pr(>|z|)
(Intercept) -4.188286233 1.94605597 -2.1521921 0.031382230 *
Overall -0.013407201 0.06158168 -0.2177141 0.827651866
RTN -0.052959314 0.05015013 -1.0560154 0.290961160
Recorded 0.162863294 0.07290053 2.2340482 0.025479900 *
PV -0.086743611 0.02950620 -2.9398438 0.003283778 **
Expire -0.035046322 0.04577103 -0.7656878 0.443862068
Trial 0.007220173 0.03294419 0.2191637 0.826522498
Fitness 0.056135418 0.03114687 1.8022810 0.071501212 .
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 757.25 on 572 degrees of freedom
Residual deviance: 725.66 on 565 degrees of freedom
AIC: 741.66
Number of Fisher Scoring iterations: 4
What I am hoping to achieve is to get variables names and coefficients of those variables which have a *, **, or *** next to their Pr(>|z|) value. In other words, I want the aforementioned variables and coefficients with a Pr(>|z|) < .05.
Ideally, I'd like to get them in a data frame. Unfortunately, the following code I've tried does not work.
variable_try <-
summary(score)$coefficients[if(summary(score)$coefficients[, 4] <= .05,
summary(score)$coefficients[, 1]),]
Error: unexpected ',' in "variable_try <-
summary(score)$coefficients[if(summary(score)$coefficients[,4] < .05,"
What about this:
data.frame(summary(score)$coef[summary(score)$coef[,4] <= .05, 4])