Risk Management Analysis Using R: A Study on Auto Accidents

1 2-1 Module Two Exercise: Introduction to Risk Management Analyses Using R Anastasia Bolinaga Southern New Hampshire University DAT 610: Optimization and Risk Assessment Professor Kyle Camac March 10, 2024

2 Introduction Company XYZ wishes to utilize Data analytics in order to assist in the process of identifying and managing risks. It was decided to study the Auto Accident Personal Injury Claim Data Set file first in order to start the investigation. To accomplish this, a formal review of Auto Accident Personal Injury Claim file will be completed using R (SNHU DAT-610-Module Two, 2024). Data preprocessing 1) Open Auto Accident Personal Injury Claim csv file in R and create a dataset “ModelTwo” and leave out records with missing data using a combination of read.csv() and na.omit() functions. 2) Display the structure of dataset attributes using str() function (“ R Reference card”, 2004)

3 The dataset consists of 45 attributes that are either numeric, integer or character in format. Since we need to perform calculations on CLAIM_AMOUNT and PAID_AMOUNT, character format is not appropriate for these attributes. 3) Remove “,” and “$” signs using gsub() function and assign to numeric format using as.numeric() function (“ R Reference card”, 2004).

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

4 Dataset descriptive analysis Obtain summary of the dataset (mean, median, 25 th , 75 th quartiles, min, max) with summary() function (“ R Reference card”, 2004). Since Claim_Number and Policy_ID are used for identification purposes only, and IND_01- IND_20/ RIDIT_01-RIDIT_20 are metrics analyzed to generate Claim_Suspicion_Score, we will focus on the summary of Claim_Amount, Paid_Amount and Claim_Suspicion_Score.

5 Minimum and maximum Paid_Amounts are lower than Claim_Amounts which is in line with the expectations of not approving 100% of all claims. The Mean is greater than the Median in both attributes indicating a right skewed distribution. 1 st and 3 rd quartile, median and mean of the Claim_Suspicion_Score is roughly the same indicating that the majority of claims are in between not being suspicious (1) to being very suspicious (5), which does not help with flagging claims for the SIU processing. Aligning Claims File to the Seven Basel II Risk Categories Basel II describes seven risk categories: (1) Internal Fraud, (2) External Fraud, (3) Employment Practices and Workplace Safety, (4) Clients, Products, and Business Practices, (5) Damage to Physical Assets, (6) Business Disruption and System Failures and finally, (7) Execution, Delivery and Process Management (Girling, 2022). Risk Category Definition Event Examples XYZ claim file Internal Fraud “Losses due to acts of a type intended to defraud, misappropriate property or circumvent regulations, the law, or company policy, excluding diversity/discrimination events, which involves at least one internal party.” (Girling, 2022) Incorrect settlements, bribery, rogue trading, etc. (Prorokowski, 2015) Always a risk due to human involvement. Currently, there is no link to staff assigned to work the claim in the dataset. External Fraud “Losses due to acts of a Fraudulent activity An example could be

6 type intended to defraud, misappropriate property, or circumvent the law, by a third party.” (Girling, 2022) by a supplier, fraudulent external access to systems (Prorokowski, 2015) purposefully causing an accident to be compensated by the insured. Potentially captured within IND_01-IND_20 metrics and summarized in Claim_Suspicion_Score Employment Practices and Workplace Safety “Losses arising from acts inconsistent with employment, health, or safety laws or agreements; from payment of personal injury claims; or from diversity/discrimination events.” (Girling, 2022) Contractual disputes with employees, departure of the key staff, discrimination claims, etc. (Prorokowski, 2015) SIU is understaffed which can potentially result in burnout and potential departure of key staff. Currently, there is no link to staff assigned to work the claim in the dataset to gauge the workload and average turnaround times. Clients, Products, and Business Practices “Losses arising from an unintentional or negligent failure to meet a professional obligation to specific clients (including fiduciary and suitability requirements), or from the nature or design of a product” (Girling, Clients’ complaints against the insurance company, billing not in accordance with agreed rates, failure in the regulatory reporting, failure in the development, design, implementation or Potential scenario is when the company cannot not fulfill their fiduciary duty to the insured as part of the agreement within the policy due to a negligence. The dataset does not include information on claim accuracy.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

7 2022). documentation of a new product (Prorokowski, 2015) Damage to Physical Assets “Losses arising from loss or damage to physical assets from natural disaster or other events” (Girling, 2022). Damage to critical buildings due to fire, flood or other natural disaster (Prorokowski, 2015) Always a risk. Impossible to predict using operational data. Business Disruption and System Failures “Losses arising from disruption of business or system failures” (Girling, 2022). System freezes, no connectivity, application glitch, missing software update, system outage due to failure of internal firewalls (Prorokowski, 2015) Claim Handling application glitch can result in data loss. Current version of the claim file has a number of records with missing data potentially due to system failures. Execution, Delivery, and Process Management “Losses from failed transaction processing or process management, from relations with trade counterparties and vendors” (Girling, 2022) Breach of limit, trading error, entering “accept” instead of “reject”, Duplicated payment of a claim (Prorokowski, 2015) There is a potential to analyze the existing dataset for duplicates (i.e. claims paid twice) or payments over policy limits. R Functionality to Operations Risk Framework

8 For XYZ, manually verifying each claim for fraud detection is impractical due to the large number of insurance claims. As a result, they can benefit from data mining and machine learning techniques that are frequently employed to identify fraud across a range of industries. The summary analysis performed above is one of the more basic analytical techniques in R. At the same time, R is also capable of performing complex operations, such as building a decision tree model to assist in identifying the likelihood of an event occurring. A Regression Model, Support Vector Machine or a Naïve Bayes model are largely used to examine auto insurance fraud. R has the capabilities of employing such techniques for fraud detection (Aslam et al., 2022).

9 References Aslam, F., Hunjra, A. I., Ftiti, Z., Louhichi, W., & Shams, T. (2022). Insurance fraud detection: Evidence from artificial intelligence and machine learning. Research in International Business and Finance , 62 , 101744. https://doi.org/10.1016/j.ribaf.2022.101744 Girling, P. X. (2022). Operational risk management: A Complete Guide for Banking and Fintech . John Wiley & Sons. Prorokowski, L. (2015), " Operational risk capital charges (Basel II): factoring in external loss data to the internal datasets ", Journal of Risk Finance , Vol. 16 No. 5, pp. 519- 535. https://doi-org.ezproxy.snhu.edu/10.1108/JRF-05-2015-0049 R Reference card (2004, November 7). https://cran.r-project.org/doc/contrib/Short-refcard.pdf SNHU DAT-610-Module Two (2024, March). 2-1 Module Two Exercise: Introduction to Risk Management Analyses Using R . https://learn.snhu.edu/d2l/le/content/1504594/viewContent/29838891/View

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

2-1 Module Two Exercise_Introduction to Risk Management Analyses Using R

Related Documents