EXERCISE 3 AND 4
# Week 3 & 4 # Step 1: Install package dplyr & read it, if you install it for the week 2 , you do not need to install again install.packages('dplyr') library(dplyr)
# Step 2: Read your data # Pl change the location of file hosp <- read.csv("C:/Users/hzare/Dropbox/HMGT400HOSPITAL.csv", header=T, sep = ',')
#Step 3: See the variables' names names (hosp)
#step 4 : generate benefit variable hosp$benefit <- (hosp$total_hosp_revenue-hosp$total_hosp_cost)
#step 5 : Define herf_cat as a categorical variable herfindahl <- group_by(hosp, herf_cat)
# Step 6: See the means # Note: 0=High Competevie # 1=Moderate Competevie # 2=low Competevie
summarize (herfindahl, bed=mean(hospital_beds, na.rm=T), payer=mean(total_hospital_employees_on_payr, na.rm=T), nopayer=mean(total_hospital_non_paid_workers, na.rm=T), resident=mean(interns_and_residents, na.rm=T), member=mean(system_member, na.rm=T), cost=mean(total_hosp_cost, na.rm=T), revenue=mean(log_hosp_revenue, na.rm=T), benefit=mean(benefit, na.rm=T), medicare=mean(total_hospital_medicare_days, na.rm=T), mediciad=mean(total_hospital_medicaid_days, na.rm=T), totdis=mean(total_hospital_discharges, na.rm=T), mediciaredis=mean(total_hospital_medicare_discharg, na.rm=T), mediciaddis=mean(total_hospital_medicaid_discharg, na.rm=T), herf_index=mean(herf_index, na.rm=T))
# Step 7: See the SD summarize (herfindahl, bed=sd(hospital_beds, na.rm=T), payer=sd(total_hospital_employees_on_payr, na.rm=T), nopayer=sd(total_hospital_non_paid_workers, na.rm=T), resident=sd(interns_and_residents, na.rm=T), member=sd(system_member, na.rm=T), cost=sd(total_hosp_cost, na.rm=T), revenue=sd(log_hosp_revenue, na.rm=T), benefit=sd(benefit, na.rm=T), medicare=sd(total_hospital_medicare_days, na.rm=T), mediciad=sd(total_hospital_medicaid_days, na.rm=T), totdis=sd(total_hospital_discharges, na.rm=T), mediciaredis=sd(total_hospital_medicare_discharg, na.rm=T), mediciaddis=sd(total_hospital_medicaid_discharg, na.rm=T), herf_index=sd(herf_index, na.rm=T))
# To find the N use following commands # Step 8: Generate 2 dataset for a ttest. high_compt <- subset(hosp, hosp$herf_cat==0) moderate_compt <- subset(hosp, hosp$herf_cat==1) low_compt <- subset(hosp, hosp$herf_cat==2)
#8-1 number of beds in high competetive market mytable <- table(high_compt$hospital_beds) summary(mytable)
#8-2 number of beds in moderate competetive market mytable <- table(moderate_compt$hospital_beds) summary(mytable)
#8-3 number of beds in low competetive market mytable <- table(low_compt$hospital_beds) summary(mytable)
#Note 2: For other variables just replace the name of variables with hospital_beds
#9 The exercise asks you to report Anova, you can report chi-squire as well #9A This is the test for chi-sq bed = table(hosp$hospital_beds, hosp$herf_cat) chisq.test(bed)
# Ignore warning message after running chi-sq
#9B This is the test for Anova bed =lm(hosp$hospital_beds ~ hosp$herf_cat) anova(bed)
#
Week
3
&
4
#
Step
1:
Install
package
dplyr
&
read
it,
if
you
install
it
for
the
week
2
,
you
do
not
need
to
install
again
install.packages('dplyr')
library(dplyr)
#
Step
2:
Read
your
data
#
Pl
change
the
location
of
file
hosp
<
-
read.csv("C:/Users/hzare/Dropbox/
HMGT400HOSPITAL.csv",
header=T,
sep
=
',')
#Step
3:
See
the
variables'
names
names
(hosp)
#step
4
:
generate
benefit
variable
hosp$benefit
<
-
(hosp$total_hosp_revenue
-
hosp$total_hosp_cost)
#step
5
:
Define
herf_cat
as
a
categorical
variable
herfindahl
<
-
group_by(hosp,
herf_cat)
#
Step
6:
See
the
means
#
Note:
0=High
Competevie
#
1=Moderate
Competevie
#
2=low
Competevie
summarize
(herfindahl,
bed=mean(hospital_beds,
na.rm=T),
payer=mean(total_hospital_employees_on_payr,
na.rm=T),
nopayer=mean(total_hospital_non_paid_workers,
na.rm=T),
resident=mean(interns_and_residents,
na.rm=T),
member=mean(system_member,
na.rm=T),
cost=mean(total_hosp_cost,
na.rm=T),
revenue=mean
(log_hosp_revenue,
na.rm=T),
benefit=mean(benefit,
na.rm=T),
medicare=mean(total_hospital_medicare_days,
na.rm=T),
mediciad=mean(total_hospital_medicaid_days,
na.rm=T),
totdis=mean(total_h
ospital_discharges,
na.rm=T),
mediciaredis=mean(total_hospital_medicare_discharg,
na.rm=T),
mediciaddis=mean(total_hospital_medicaid_discharg,
na.rm=T),
herf_index=mean(herf_index,
na.rm=T))
#
Step
7:
See
the
SD
summarize
(herfindahl,
bed=sd(hospital_beds,
na.rm=T),
payer=sd(total_hospital_employees_on_payr,
na.rm=T),
nopayer=sd(total_hospital_non_paid_workers,
na.rm=T),
resident=sd(interns_and_residents,
na.rm=T),
mem
ber=sd(system_member,
na.rm=T),
cost=sd(total_hosp_cost,
na.rm=T),
revenue=sd(log_hosp_revenue,
na.rm=T),
benefit=sd(benefit,
na.rm=T),
medicare=sd(total_hospital_medicare_days,
na.rm=T),
mediciad=sd(total_hospital_medicaid_days,
na.rm=T),
totdis=sd(total_hospital_discharges,
na.rm=T),
mediciaredis=sd(total_hospital_medicare_discharg,
na.rm=T),
mediciaddis=sd(total_hospital_medicaid_di
scharg,
na.rm=T),
# Week 3 & 4
# Step 1: Install package dplyr & read it, if you install it for the week 2 ,
you do not need to install again
install.packages('dplyr')
library(dplyr)
# Step 2: Read your data
# Pl change the location of file
hosp <- read.csv("C:/Users/hzare/Dropbox/HMGT400HOSPITAL.csv", header=T, sep
= ',')
#Step 3: See the variables' names
names (hosp)
#step 4 : generate benefit variable
hosp$benefit <- (hosp$total_hosp_revenue-hosp$total_hosp_cost)
#step 5 : Define herf_cat as a categorical variable
herfindahl <- group_by(hosp, herf_cat)
# Step 6: See the means
# Note: 0=High Competevie
# 1=Moderate Competevie
# 2=low Competevie
summarize (herfindahl, bed=mean(hospital_beds, na.rm=T),
payer=mean(total_hospital_employees_on_payr, na.rm=T),
nopayer=mean(total_hospital_non_paid_workers, na.rm=T),
resident=mean(interns_and_residents, na.rm=T),
member=mean(system_member, na.rm=T),
cost=mean(total_hosp_cost, na.rm=T),
revenue=mean(log_hosp_revenue, na.rm=T),
benefit=mean(benefit, na.rm=T),
medicare=mean(total_hospital_medicare_days, na.rm=T),
mediciad=mean(total_hospital_medicaid_days, na.rm=T),
totdis=mean(total_hospital_discharges, na.rm=T),
mediciaredis=mean(total_hospital_medicare_discharg, na.rm=T),
mediciaddis=mean(total_hospital_medicaid_discharg, na.rm=T),
herf_index=mean(herf_index, na.rm=T))
# Step 7: See the SD
summarize (herfindahl, bed=sd(hospital_beds, na.rm=T),
payer=sd(total_hospital_employees_on_payr, na.rm=T),
nopayer=sd(total_hospital_non_paid_workers, na.rm=T),
resident=sd(interns_and_residents, na.rm=T),
member=sd(system_member, na.rm=T),
cost=sd(total_hosp_cost, na.rm=T),
revenue=sd(log_hosp_revenue, na.rm=T),
benefit=sd(benefit, na.rm=T),
medicare=sd(total_hospital_medicare_days, na.rm=T),
mediciad=sd(total_hospital_medicaid_days, na.rm=T),
totdis=sd(total_hospital_discharges, na.rm=T),
mediciaredis=sd(total_hospital_medicare_discharg, na.rm=T),
mediciaddis=sd(total_hospital_medicaid_discharg, na.rm=T),