LMS: Least Minimum Sum

Description Usage Arguments Details Value Examples

View source: R/minimum_sum_method.R

Description

"LMS = sum of the ne sample overpayments + sum of the smallest LNE - ne nonsampled population payments"

Usage

1
LMS(df_samp, df_frame, var, sub_var, flag, Npop, nsamp, ne, alpha = 0.1)

Arguments

df_samp

Sample (must have var, flag, and sub_var)

df_frame

sampling frame (must have sub_var variable to match with df_samp)

var

Independent variable in LMS calculation

sub_var

Variable used to match sample records (df_samp) and sampling frame records (df_frame)

flag

Variable that flags observations with errors.

Npop

"Number of payments (e.g. on Medicare claims) in the universe/population"

nsamp

"Number of payments (e.g. on Medicare claims) in...simple random sample"

ne

"The number...of sample payments which are completely in error (Or, for partial overpayment scenarios, seriously in error—see Section 4.2.)"

alpha

alpha-level for "1-alpha confidence bound"

Details

Citation: Edwards, D., Ward-Besser, G., Lasecki, J., Parker, B., Wu, F. & Moorhead, P. (2003). The Minimum Sum Method: A Distribution-Free Sampling Procedure for Medicare Fraud Investigations. Heath Services & Outcomes Research Methodology 4: 241-263.

Value

LMS nested list returning a list of numbers and a list of data_frames

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
#Generate sampling frame  with an index id from 1 to N
df_sample_frame_num <- data.frame("sample_frame_sequence_id" = 1:1000,
                                 "score" = rnorm(1000, 75, 10))

#Get output and input in lists
score_audit_num <- rs_singlestage(df = df_sample_frame_num,
                                  audit_review = "Score Audit",
                                  quantity_to_generate = 100,
                                  quantity_of_spares = 3,
                                  frame_low = 1,
                                  frame_high = 1000)

score_samp <- score_audit_num$output$sample
score_samp$audit_results <- sample(0:1, 100, replace = TRUE)

score_frame <- score_audit_num$output$sample_frame

LMS_out <- LMS(score_samp, score_frame, var = score, sub_var = sample_frame_sequence_id,
               flag = audit_results,
               Npop = 1000, nsamp = 100, ne = 100)

#Get LMS estimate
LMS_out$numbers$LMS

mncube/claimr documentation built on Dec. 21, 2021, 8:07 p.m.