Brad
Sith
3/4/2023
Graphing Functions
require(ggplot2)## Loading required package: ggplot2## Warning: package 'ggplot2' was built under R version 4.2.2require(ggcorrplot)## Loading required package: ggcorrplot## Warning: package 'ggcorrplot' was built under R version 4.2.2corfunction=function(d){
mycorr=cor(d[, 1:ncol(d)]); p.mat=ggcorrplot::cor_pmat(d[,1:ncol(d)])
myplot=ggcorrplot(mycorr, hc.order=TRUE,type="lower",
colors=c("red", "white","green"),tl.cex = 8,
tl.col = "black", lab=TRUE, lab_size=2, p.mat=p.mat,
insig="pch", pch=4)
print(myplot)}Import Data
require(foreign)## Loading required package: foreignrequire(psych)## Loading required package: psych## Warning: package 'psych' was built under R version 4.2.2##
## Attaching package: 'psych'## The following objects are masked from 'package:ggplot2':
##
## %+%, alpharequire(Amelia)## Loading required package: Amelia## Warning: package 'Amelia' was built under R version 4.2.2## Loading required package: Rcpp## ##
## ## Amelia II: Multiple Imputation
## ## (Version 1.8.0, built: 2021-05-26)
## ## Copyright (C) 2005-2023 James Honaker, Gary King and Matthew Blackwell
## ## Refer to http://gking.harvard.edu/amelia/ for more information
## ##mydata=read.spss('c:/users/lfult/downloads/brad.sav', to.data.frame=T)
mydata$DEPENDENT_VARIABLES=mydata$INDEPENDENT_VARIABLES=NULL
mydata$Region=as.factor(mydata$C_Region2*2+
mydata$C_Region3*3+mydata$C_Region4*4+
mydata$C_Region5*5+mydata$C_Region6*6+
mydata$C_Region7*7+mydata$C_Region8*8+ mydata$C_Region9*9)
mydata$C_Region2=mydata$C_Region3=mydata$C_Region4=mydata$C_Region5=mydata$C_Region6=mydata$C_Region7=mydata$C_Region8=mydata$C_Region9=NULL
mydata$Region=addNA(mydata$Region, 0)Reformat Factors
mydata$PROVIDER_NUMBER=NULL
mydata$MCI=as.factor(round(mydata$MARKET_CONCENT_INDEX,0))
mydata$MARKET_CONCENT_INDEX=NULL
mydata$Type=as.factor(mydata$SOLE_COMMUNITY_HOSPITAL+mydata$FOR_PROFIT*2+mydata$GOVT_OPERATED*3)
mydata$Urban=as.factor(mydata$GEOGRAPHIC_CLASSIFICATION_Urban_0)
mydata$Region=as.factor(mydata$Region)
mydata$FOR_PROFIT=mydata$SOLE_COMMUNITY_HOSPITAL=mydata$GOVT_OPERATED=
mydata$GEOGRAPHIC_CLASSIFICATION_Urban_0=NULLIdentify Missing by Column
myc = function(x){
co=rep(0,ncol(mydata))
for (i in 1:ncol(x)){co[i]=sum(is.na(x[1:nrow(x), i]))}
names(co)=colnames(mydata)
co=sort(co, decreasing=T)/nrow(mydata)
print(length(co[co>.2]))
tmp=co[1:13]
barplot(tmp, las=2, cex.names=.5, space=0)
print(names(tmp))
return(co)
}
myc(mydata)## [1] 1
## [1] "TOTAL_PERFORMANCE_SCORE"
## [2] "DEBT_TO_EQUITY_RATIO"
## [3] "MCI"
## [4] "CHARITY_CARE_COSTS_Scaled"
## [5] "PERCENET_MEDICAID_DAYS"
## [6] "SERIOUS_COMPLICATION_RATE"
## [7] "PERCENT_MEDICARE_DAYS"
## [8] "AVG_AGE_FACILITY"
## [9] "BAD_DEBT_NPR_RATIO"
## [10] "Type"
## [11] "TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled"
## [12] "CASE_MIX_INDEX"
## [13] "CC_MCC_RATE"## TOTAL_PERFORMANCE_SCORE
## 0.23100872
## DEBT_TO_EQUITY_RATIO
## 0.17714819
## MCI
## 0.16625156
## CHARITY_CARE_COSTS_Scaled
## 0.14601494
## PERCENET_MEDICAID_DAYS
## 0.14601494
## SERIOUS_COMPLICATION_RATE
## 0.14290162
## PERCENT_MEDICARE_DAYS
## 0.14103362
## AVG_AGE_FACILITY
## 0.13075965
## BAD_DEBT_NPR_RATIO
## 0.10554172
## Type
## 0.10149440
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled
## 0.10118306
## CASE_MIX_INDEX
## 0.10087173
## CC_MCC_RATE
## 0.10087173
## PX_REV_PER_DISCHARGE_Scaled
## 0.09869240
## BED_UTILIZATION
## 0.09869240
## AVG_LENGTH_OF_STAY
## 0.09869240
## NET_OPERATING_PROFIT_MARGIN
## 0.09806974
## RETURN_ON_ASSETS
## 0.09806974
## NPR_PER_STAFFED_BED
## 0.09806974
## EBITDA_PER_STAFFED_BED
## 0.09806974
## NET_INCOME_PER_STAFFED_BED
## 0.09806974
## LABOR_COMP_RATIO
## 0.09806974
## STAFFED_BEDS_SCALED
## 0.09806974
## Urban
## 0.09806974
## Region
## 0.00000000Just For Now, Delete Missing
mydata$AVG_AGE_FACILITY=NULL #not interesting in models
mydata=na.omit(mydata)
corfunction(mydata[,c(1:20)])
describe(mydata)## vars n mean sd median
## NET_OPERATING_PROFIT_MARGIN 1 2043 0.00 0.16 0.01
## RETURN_ON_ASSETS 2 2043 0.06 0.62 0.06
## PX_REV_PER_DISCHARGE_Scaled 3 2043 0.03 0.01 0.03
## NPR_PER_STAFFED_BED 4 2043 1.46 0.73 1.33
## EBITDA_PER_STAFFED_BED 5 2043 0.09 0.25 0.08
## NET_INCOME_PER_STAFFED_BED 6 2043 0.11 0.20 0.08
## DEBT_TO_EQUITY_RATIO 7 2043 0.86 18.16 0.22
## LABOR_COMP_RATIO 8 2043 0.43 0.12 0.42
## TOTAL_PERFORMANCE_SCORE 9 2043 33.11 10.86 32.00
## SERIOUS_COMPLICATION_RATE 10 2043 0.89 0.18 0.87
## CHARITY_CARE_COSTS_Scaled 11 2043 9.40 15.72 4.44
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled 12 2043 23.82 35.83 13.29
## BAD_DEBT_NPR_RATIO 13 2043 0.08 0.09 0.05
## BED_UTILIZATION 14 2043 0.54 0.18 0.56
## STAFFED_BEDS_SCALED 15 2043 0.24 0.22 0.18
## CASE_MIX_INDEX 16 2043 1.68 0.28 1.64
## CC_MCC_RATE 17 2043 0.02 0.01 0.02
## PERCENT_MEDICARE_DAYS 18 2043 0.13 0.06 0.13
## PERCENET_MEDICAID_DAYS 19 2043 0.04 0.04 0.03
## AVG_LENGTH_OF_STAY 20 2043 4.38 0.90 4.28
## Region* 21 2010 5.17 2.42 5.00
## MCI* 22 2043 1.30 0.46 1.00
## Type* 23 2043 1.95 1.22 1.00
## Urban* 24 2043 1.05 0.22 1.00
## trimmed mad min max
## NET_OPERATING_PROFIT_MARGIN 0.01 0.11 -1.64 0.54
## RETURN_ON_ASSETS 0.06 0.09 -25.79 3.10
## PX_REV_PER_DISCHARGE_Scaled 0.03 0.01 0.00 0.19
## NPR_PER_STAFFED_BED 1.39 0.58 0.01 11.45
## EBITDA_PER_STAFFED_BED 0.09 0.16 -2.38 2.24
## NET_INCOME_PER_STAFFED_BED 0.10 0.13 -1.43 2.27
## DEBT_TO_EQUITY_RATIO 0.31 0.39 -167.52 696.22
## LABOR_COMP_RATIO 0.42 0.12 0.15 0.98
## TOTAL_PERFORMANCE_SCORE 32.37 10.19 6.00 76.67
## SERIOUS_COMPLICATION_RATE 0.88 0.16 0.44 1.94
## CHARITY_CARE_COSTS_Scaled 6.15 5.04 0.00 228.31
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled 16.87 11.94 0.43 539.59
## BAD_DEBT_NPR_RATIO 0.06 0.04 0.00 0.95
## BED_UTILIZATION 0.55 0.20 0.07 1.00
## STAFFED_BEDS_SCALED 0.20 0.14 0.01 2.73
## CASE_MIX_INDEX 1.66 0.24 0.94 3.62
## CC_MCC_RATE 0.02 0.01 0.00 0.21
## PERCENT_MEDICARE_DAYS 0.13 0.05 0.00 0.41
## PERCENET_MEDICAID_DAYS 0.03 0.03 0.00 0.62
## AVG_LENGTH_OF_STAY 4.33 0.81 1.00 8.40
## Region* 5.13 2.97 1.00 10.00
## MCI* 1.25 0.00 1.00 2.00
## Type* 1.77 0.00 1.00 5.00
## Urban* 1.00 0.00 1.00 2.00
## range skew kurtosis se
## NET_OPERATING_PROFIT_MARGIN 2.18 -2.02 14.87 0.00
## RETURN_ON_ASSETS 28.89 -36.37 1525.57 0.01
## PX_REV_PER_DISCHARGE_Scaled 0.19 2.55 15.01 0.00
## NPR_PER_STAFFED_BED 11.44 2.51 20.21 0.02
## EBITDA_PER_STAFFED_BED 4.61 -0.70 14.76 0.01
## NET_INCOME_PER_STAFFED_BED 3.70 1.48 14.65 0.00
## DEBT_TO_EQUITY_RATIO 863.74 28.14 1069.83 0.40
## LABOR_COMP_RATIO 0.83 0.88 1.10 0.00
## TOTAL_PERFORMANCE_SCORE 70.67 0.69 0.61 0.24
## SERIOUS_COMPLICATION_RATE 1.50 1.03 2.25 0.00
## CHARITY_CARE_COSTS_Scaled 228.31 5.32 45.43 0.35
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled 539.16 5.68 48.66 0.79
## BAD_DEBT_NPR_RATIO 0.95 3.48 18.73 0.00
## BED_UTILIZATION 0.93 -0.19 -0.60 0.00
## STAFFED_BEDS_SCALED 2.72 3.01 17.62 0.00
## CASE_MIX_INDEX 2.68 1.00 2.69 0.01
## CC_MCC_RATE 0.21 4.54 39.68 0.00
## PERCENT_MEDICARE_DAYS 0.40 0.79 1.15 0.00
## PERCENET_MEDICAID_DAYS 0.62 3.15 22.28 0.00
## AVG_LENGTH_OF_STAY 7.40 0.65 1.19 0.02
## Region* 9.00 0.14 -0.97 0.05
## MCI* 1.00 0.88 -1.22 0.01
## Type* 4.00 0.87 -0.61 0.03
## Urban* 1.00 4.11 14.87 0.00Min-Max Scale and Make Positive Definite
myf=function(x)
{
x=(x-min(x))/(max(x)-min(x))
x=x+.01
return(x)
}
newdata=mydata
newdata[,1:19]=as.data.frame(lapply(newdata[,1:18],myf))
describe(newdata)## vars n mean sd median
## NET_OPERATING_PROFIT_MARGIN 1 2043 0.76 0.07 0.77
## RETURN_ON_ASSETS 2 2043 0.90 0.02 0.90
## PX_REV_PER_DISCHARGE_Scaled 3 2043 0.15 0.07 0.13
## NPR_PER_STAFFED_BED 4 2043 0.14 0.06 0.13
## NET_INCOME_PER_STAFFED_BED 5 2043 0.43 0.05 0.42
## DEBT_TO_EQUITY_RATIO 6 2043 0.20 0.02 0.20
## LABOR_COMP_RATIO 7 2043 0.35 0.15 0.33
## TOTAL_PERFORMANCE_SCORE 8 2043 0.39 0.15 0.38
## SERIOUS_COMPLICATION_RATE 9 2043 0.31 0.12 0.30
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled 10 2043 0.05 0.07 0.03
## BAD_DEBT_NPR_RATIO 11 2043 0.10 0.10 0.07
## BED_UTILIZATION 12 2043 0.52 0.19 0.54
## STAFFED_BEDS_SCALED 13 2043 0.09 0.08 0.07
## CASE_MIX_INDEX 14 2043 0.28 0.10 0.27
## CC_MCC_RATE 15 2043 0.10 0.07 0.09
## PERCENT_MEDICARE_DAYS 16 2043 0.33 0.14 0.32
## PERCENET_MEDICAID_DAYS 17 2043 0.07 0.07 0.05
## AVG_LENGTH_OF_STAY 18 2043 0.47 0.12 0.45
## Region 19 2043 0.76 0.07 0.77
## MCI* 20 2043 1.30 0.46 1.00
## Type* 21 2043 1.95 1.22 1.00
## Urban* 22 2043 1.05 0.22 1.00
## trimmed mad min max range
## NET_OPERATING_PROFIT_MARGIN 0.77 0.05 0.01 1.01 1
## RETURN_ON_ASSETS 0.90 0.00 0.01 1.01 1
## PX_REV_PER_DISCHARGE_Scaled 0.14 0.05 0.01 1.01 1
## NPR_PER_STAFFED_BED 0.13 0.05 0.01 1.01 1
## NET_INCOME_PER_STAFFED_BED 0.42 0.04 0.01 1.01 1
## DEBT_TO_EQUITY_RATIO 0.20 0.00 0.01 1.01 1
## LABOR_COMP_RATIO 0.34 0.14 0.01 1.01 1
## TOTAL_PERFORMANCE_SCORE 0.38 0.14 0.01 1.01 1
## SERIOUS_COMPLICATION_RATE 0.30 0.11 0.01 1.01 1
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled 0.04 0.02 0.01 1.01 1
## BAD_DEBT_NPR_RATIO 0.08 0.04 0.01 1.01 1
## BED_UTILIZATION 0.53 0.21 0.01 1.01 1
## STAFFED_BEDS_SCALED 0.08 0.05 0.01 1.01 1
## CASE_MIX_INDEX 0.28 0.09 0.01 1.01 1
## CC_MCC_RATE 0.09 0.04 0.01 1.01 1
## PERCENT_MEDICARE_DAYS 0.32 0.13 0.01 1.01 1
## PERCENET_MEDICAID_DAYS 0.06 0.04 0.01 1.01 1
## AVG_LENGTH_OF_STAY 0.46 0.11 0.01 1.01 1
## Region 0.77 0.05 0.01 1.01 1
## MCI* 1.25 0.00 1.00 2.00 1
## Type* 1.77 0.00 1.00 5.00 4
## Urban* 1.00 0.00 1.00 2.00 1
## skew kurtosis se
## NET_OPERATING_PROFIT_MARGIN -2.02 14.87 0.00
## RETURN_ON_ASSETS -36.37 1525.57 0.00
## PX_REV_PER_DISCHARGE_Scaled 2.55 15.01 0.00
## NPR_PER_STAFFED_BED 2.51 20.21 0.00
## NET_INCOME_PER_STAFFED_BED 1.48 14.65 0.00
## DEBT_TO_EQUITY_RATIO 28.14 1069.83 0.00
## LABOR_COMP_RATIO 0.88 1.10 0.00
## TOTAL_PERFORMANCE_SCORE 0.69 0.61 0.00
## SERIOUS_COMPLICATION_RATE 1.03 2.25 0.00
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled 5.68 48.66 0.00
## BAD_DEBT_NPR_RATIO 3.48 18.73 0.00
## BED_UTILIZATION -0.19 -0.60 0.00
## STAFFED_BEDS_SCALED 3.01 17.62 0.00
## CASE_MIX_INDEX 1.00 2.69 0.00
## CC_MCC_RATE 4.54 39.68 0.00
## PERCENT_MEDICARE_DAYS 0.79 1.15 0.00
## PERCENET_MEDICAID_DAYS 3.15 22.28 0.00
## AVG_LENGTH_OF_STAY 0.65 1.19 0.00
## Region -2.02 14.87 0.00
## MCI* 0.88 -1.22 0.01
## Type* 0.87 -0.61 0.03
## Urban* 4.11 14.87 0.00Box Cox Transformations
mymat=as.matrix(newdata[, 1:18])
require(car)## Loading required package: car## Loading required package: carData##
## Attaching package: 'car'## The following object is masked from 'package:psych':
##
## logitmybox=powerTransform(mymat~1)
mybox## Estimated transformation parameters
## NET_OPERATING_PROFIT_MARGIN
## 3.25660544
## RETURN_ON_ASSETS
## 7.44016256
## PX_REV_PER_DISCHARGE_Scaled
## 0.17678801
## NPR_PER_STAFFED_BED
## 0.19508175
## NET_INCOME_PER_STAFFED_BED
## 0.82278724
## DEBT_TO_EQUITY_RATIO
## 0.36668676
## LABOR_COMP_RATIO
## 0.40548949
## TOTAL_PERFORMANCE_SCORE
## 0.53790342
## SERIOUS_COMPLICATION_RATE
## 0.44274518
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled
## -0.32292914
## BAD_DEBT_NPR_RATIO
## -0.11951089
## BED_UTILIZATION
## 0.41346084
## STAFFED_BEDS_SCALED
## -0.07777543
## CASE_MIX_INDEX
## 0.40948035
## CC_MCC_RATE
## 0.04298725
## PERCENT_MEDICARE_DAYS
## 0.43794168
## PERCENET_MEDICAID_DAYS
## -0.12806952
## AVG_LENGTH_OF_STAY
## 0.43571481testTransform(mybox,mybox$lambda)## LRT
## LR test, lambda = (3.26 7.44 0.18 0.2 0.82 0.37 0.41 0.54 0.44 -0.32 -0.12 0.41 -0.08 0.41 0.04 0.44 -0.13 0.44) 0
## df
## LR test, lambda = (3.26 7.44 0.18 0.2 0.82 0.37 0.41 0.54 0.44 -0.32 -0.12 0.41 -0.08 0.41 0.04 0.44 -0.13 0.44) 18
## pval
## LR test, lambda = (3.26 7.44 0.18 0.2 0.82 0.37 0.41 0.54 0.44 -0.32 -0.12 0.41 -0.08 0.41 0.04 0.44 -0.13 0.44) 1require(ResourceSelection)## Loading required package: ResourceSelection## ResourceSelection 0.3-5 2019-07-22for (i in 1:18)
{
newdata[,i]=newdata[,i]^mybox$lambda[i]
}Kernel Density Estimates / Pairs Plots
kdepairs(newdata[,c(1:9)])## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
kdepairs(newdata[,c(10:18)])## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
## Warning in par(usr): argument 1 does not name a graphical parameter
Initial Regression After StepAIC (Note the Outliers)
library(kableExtra)## Warning: package 'kableExtra' was built under R version 4.2.2## Warning in !is.null(rmarkdown::metadata$output) && rmarkdown::metadata$output
## %in% : 'length(x) = 2 > 1' in coercion to 'logical(1)'library(MASS)
mylm=lm(PX_REV_PER_DISCHARGE_Scaled ~RETURN_ON_ASSETS + NPR_PER_STAFFED_BED +
NET_INCOME_PER_STAFFED_BED + LABOR_COMP_RATIO + SERIOUS_COMPLICATION_RATE +
TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled + BAD_DEBT_NPR_RATIO +
BED_UTILIZATION + STAFFED_BEDS_SCALED + CASE_MIX_INDEX +
CC_MCC_RATE + PERCENT_MEDICARE_DAYS + PERCENET_MEDICAID_DAYS +
AVG_LENGTH_OF_STAY + Region + Type, data=newdata)
summary(mylm)##
## Call:
## lm(formula = PX_REV_PER_DISCHARGE_Scaled ~ RETURN_ON_ASSETS +
## NPR_PER_STAFFED_BED + NET_INCOME_PER_STAFFED_BED + LABOR_COMP_RATIO +
## SERIOUS_COMPLICATION_RATE + TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled +
## BAD_DEBT_NPR_RATIO + BED_UTILIZATION + STAFFED_BEDS_SCALED +
## CASE_MIX_INDEX + CC_MCC_RATE + PERCENT_MEDICARE_DAYS + PERCENET_MEDICAID_DAYS +
## AVG_LENGTH_OF_STAY + Region + Type, data = newdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.09402 -0.00494 0.00003 0.00482 0.38407
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) -1.071e-02 2.007e-02 -0.534
## RETURN_ON_ASSETS 2.363e-02 1.090e-02 2.167
## NPR_PER_STAFFED_BED 1.032e+00 9.247e-03 111.601
## NET_INCOME_PER_STAFFED_BED -1.749e-02 7.843e-03 -2.230
## LABOR_COMP_RATIO 8.782e-03 3.788e-03 2.319
## SERIOUS_COMPLICATION_RATE 6.527e-03 2.990e-03 2.183
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled -3.481e-03 8.161e-04 -4.265
## BAD_DEBT_NPR_RATIO 2.163e-02 3.162e-03 6.841
## BED_UTILIZATION -3.631e-01 3.803e-03 -95.486
## STAFFED_BEDS_SCALED 2.899e-02 8.773e-03 3.305
## CASE_MIX_INDEX 3.437e-02 5.169e-03 6.649
## CC_MCC_RATE -6.307e-02 1.493e-02 -4.224
## PERCENT_MEDICARE_DAYS 1.383e-02 2.895e-03 4.777
## PERCENET_MEDICAID_DAYS 1.496e-02 2.231e-03 6.703
## AVG_LENGTH_OF_STAY 3.196e-01 5.174e-03 61.769
## Region 1.010e-02 6.007e-03 1.682
## Type1 7.002e-04 1.075e-03 0.652
## Type2 9.395e-05 8.840e-04 0.106
## Type3 3.138e-03 1.007e-03 3.116
## Type4 2.153e-03 1.843e-03 1.168
## Pr(>|t|)
## (Intercept) 0.593638
## RETURN_ON_ASSETS 0.030316 *
## NPR_PER_STAFFED_BED < 2e-16 ***
## NET_INCOME_PER_STAFFED_BED 0.025832 *
## LABOR_COMP_RATIO 0.020510 *
## SERIOUS_COMPLICATION_RATE 0.029185 *
## TOT_UNCOMPENSATED_CARE_UNREIMBURSED_COSTS_Scaled 2.09e-05 ***
## BAD_DEBT_NPR_RATIO 1.04e-11 ***
## BED_UTILIZATION < 2e-16 ***
## STAFFED_BEDS_SCALED 0.000967 ***
## CASE_MIX_INDEX 3.79e-11 ***
## CC_MCC_RATE 2.51e-05 ***
## PERCENT_MEDICARE_DAYS 1.91e-06 ***
## PERCENET_MEDICAID_DAYS 2.64e-11 ***
## AVG_LENGTH_OF_STAY < 2e-16 ***
## Region 0.092730 .
## Type1 0.514778
## Type2 0.915369
## Type3 0.001856 **
## Type4 0.242796
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.01324 on 2023 degrees of freedom
## Multiple R-squared: 0.9417, Adjusted R-squared: 0.9412
## F-statistic: 1721 on 19 and 2023 DF, p-value: < 2.2e-16#mystep=stepAIC(mylm)
hist(mylm$residuals)
par(ask=F)
plot(mylm)
