Difference between revisions of "AY1516 T2 Team Hew - Overview/Interim Review"

From Analytics Practicum
Jump to navigation Jump to search
 
(17 intermediate revisions by the same user not shown)
Line 33: Line 33:
 
<div style="margin:20px; padding-top:0px; padding-right:20px; padding-left:20px; background: #ffffff; font-family: Trebuchet MS, sans-serif; font-size: 100%; -webkit-border-radius: 10px;-webkit-box-shadow: 1px 1px 2px rgba(0, 0, 0, 0.08); -moz-box-shadow:    1px 1px 2px rgba(0, 0, 0, 0.08);box-shadow: 1px 1px 2px rgba(0, 0, 0, 0.08);">
 
<div style="margin:20px; padding-top:0px; padding-right:20px; padding-left:20px; background: #ffffff; font-family: Trebuchet MS, sans-serif; font-size: 100%; -webkit-border-radius: 10px;-webkit-box-shadow: 1px 1px 2px rgba(0, 0, 0, 0.08); -moz-box-shadow:    1px 1px 2px rgba(0, 0, 0, 0.08);box-shadow: 1px 1px 2px rgba(0, 0, 0, 0.08);">
  
== <p style="font-family:Trebuchet MS; border-left: 6px solid #62b762; padding-left:10px; line-height:40px; height:40px"><b>Project Changes </b></p>==
+
== <p style="font-family:Trebuchet MS; border-left: 6px solid #62b762; padding-left:10px; line-height:40px; height:40px"><b>Revised Project Background</b></p>==
 
Initially, it was mentioned in the proposal that Tokio Marine Life Insurance Singapore was to provide us with their data for analysis. However due to unforeseen circumstances, they were unable to extract and anonymize the data in time. After discussing with our Project Sponsor (Benito Mable), we will be focusing on another dataset which was supplied by TMI as of end January 2016. Objectives were subsequently revised with our Project Sponsor to reflect the different nature of this new dataset.
 
Initially, it was mentioned in the proposal that Tokio Marine Life Insurance Singapore was to provide us with their data for analysis. However due to unforeseen circumstances, they were unable to extract and anonymize the data in time. After discussing with our Project Sponsor (Benito Mable), we will be focusing on another dataset which was supplied by TMI as of end January 2016. Objectives were subsequently revised with our Project Sponsor to reflect the different nature of this new dataset.
 
<br/><br/>
 
<br/><br/>
Line 52: Line 52:
 
<br/><br/>
 
<br/><br/>
 
'''Example Data:'''
 
'''Example Data:'''
 +
[[File:Example data.JPG|center|Example data]]
 +
<br/>
 +
Each ''Risk_NO'' represents a vehicle insured under the same ''Policy'' number, which can be referenced to a particular customer. However, the dataset does not include any customer details.
  
 +
<br/>
 +
<br/>
 +
=== Claims Process ===
 +
The multiple ''Transaction_NO'' values for each ''Claim_NO'' serves to record the transactions corresponding to various steps in the claims-making process.
 +
<br/>
 +
[[File:Claims process.JPG|center]]
 +
<br/>
 +
The columns of ''ClaimOS'', ''ClaimPaid'' and ''ClaimInc'' emulate the relationship of that of a balance sheet, and can be represented by the equation below:
 +
<br/><br/>
 +
<center>'''<big>ClaimPaid + ClaimOS = ClaimInc</big>'''</center>
 +
<br/>
 +
Each row of these columns show the transactions applied, but does NOT give the current value of that column. To derive the actual amount of the total ''ClaimPaid'', we can simply sum up the transactions, or use the value in the very last transaction (shown by the cell highlighted in green). This is best illustrated by the example below:
 +
 +
[[File:Claims transaction.JPG|center]]
 +
<br/>
 +
''Transaction_NO'' = 1 corresponds to Step 3 in the Claims Process chart, where the ''ClaimOS'' is an estimate of the ''ClaimPaid'' recorded when Tokio Marine staff have assessed the damage. ''ClaimInc'' is subsequently 500,000 as calculated by the equation given. ''Transaction_NO'' = 2 corresponds to an adjustment in the estimated figure, to more accurately reflect the actual damage. ''Transaction_NO'' = 3 and 4 are the transactions recording the customer receiving the claim amount from Tokio Marine.
 +
<br/><br/>
 +
===Revised Limitations===
 +
{| class="wikitable"
 +
|-
 +
! Limitation !! Solution
 +
|-
 +
| Lack of customer demographic data, as the data is collected only when a claim is made || Limit certain analysis to only claims subset data
 +
|-
 +
| Lack of other costs like overhead costs which increases cost of each policy  || Compare only gross loss ratio and not combined ratio with competitors
 +
|-
 +
| Complicated business relationships which might skew analysis or have counter-intuitive results || Frequent consultations with project sponsor to clarify any surprising and counter-intuitive results
 +
|}
 
<br/><br/>
 
<br/><br/>
 
</div>
 
</div>
Line 59: Line 90:
 
<div style="margin:20px; padding-top:0px; padding-right:20px; padding-left:20px; background: #ffffff; font-family: Trebuchet MS, sans-serif; font-size: 100%; -webkit-border-radius: 10px;-webkit-box-shadow: 1px 1px 2px rgba(0, 0, 0, 0.08); -moz-box-shadow:    1px 1px 2px rgba(0, 0, 0, 0.08);box-shadow: 1px 1px 2px rgba(0, 0, 0, 0.08);">
 
<div style="margin:20px; padding-top:0px; padding-right:20px; padding-left:20px; background: #ffffff; font-family: Trebuchet MS, sans-serif; font-size: 100%; -webkit-border-radius: 10px;-webkit-box-shadow: 1px 1px 2px rgba(0, 0, 0, 0.08); -moz-box-shadow:    1px 1px 2px rgba(0, 0, 0, 0.08);box-shadow: 1px 1px 2px rgba(0, 0, 0, 0.08);">
  
== <p style="font-family:Trebuchet MS; border-left: 6px solid #daad25; padding-left:10px; line-height:40px; height:40px"><b>Objectives </b></p>==
+
== <p style="font-family:Trebuchet MS; border-left: 6px solid #daad25; padding-left:10px; line-height:40px; height:40px"><b>Revised Objectives </b></p>==
There are a few objectives which were formulated together with Tokio Marine at the initial meeting. Only some of the objectives listed down will be chosen as the focus as final paper as it is dependent on the availability and feasibility of data.
+
# Motor Insurance overall profitability and profitability by brand
 
+
# Motor Insurance and claims trends
# Develop a database analysis to formulate a demographic and psychographic profile of customers 
+
# Analyse characteristics of Top Agents by Loss Ratio and profitability
# Analyse average product holdings per customer 
+
# Marketing recommendations to improve business performance based on our findings
# Determine which customer segments and products are more profitable 
+
# Time series forecasting of profitability (which will only be discussed more fully in the final report)
# Which channels are more profitable (i.e. direct, online or through agents) 
+
# Other predictive models (e.g. Multiple Linear Regression), as a bonus
# Propensity to buy or assess next best offer for customers to enhance effectiveness of marketing campaigns. 
+
<br/>
 
 
The first objective is to demographically segment customers for more targeted marketing efforts. A psychographic profile here refers to the tendencies/behaviours of certain demographic groups. Tokio Marine will provide external datasets and research to supplement our findings. The second objective would be to investigate when and how often a certain demographic terminates/renews their policies. The third objective, linked to the first, is to determine which customer segments are more profitable. The fourth objective is to determine which sales channels are more profitable and thus where efforts should be concentrated on. The last objective is to determine the customer’s propensity to buy and also the next best offer.
 
<br/><br/>
 
 
'''Scope of Work'''<br/>
 
'''Scope of Work'''<br/>
While Tokio Marine has several objectives, we will be directing our efforts towards Objective 1 due to time constraints. This is also the main objective which Tokio Marine hopes to achieve.
+
Due to the sheer number of records in the dataset, we will be restricting our scope to only focus on policies which were underwritten starting 2012 to 2015. Product lines like Shariah/Takaful Islamic motor insurance were discontinued in 2013 so their data will be excluded from the analysis
<br/><br/>
+
<br/><br/><br/>
 
</div>
 
</div>
  
Line 78: Line 106:
  
 
== <p style="font-family:Trebuchet MS; border-left: 6px solid #cf657b; padding-left:10px; line-height:40px; height:40px"><b>Research & Methodology </b></p>==
 
== <p style="font-family:Trebuchet MS; border-left: 6px solid #cf657b; padding-left:10px; line-height:40px; height:40px"><b>Research & Methodology </b></p>==
===Review of Similar Work===
+
===Literature Review===
Typically, Affinity Analysis has been applied to consumer products like groceries and thus there exists little literature on this. However recently, Affinity Analysis has been used to analyze different customer segments and this has helped in supporting their decisions to target certain demographic groups<ref name="market-basket-analysis">Roodpishi, M & Nashtaei, R. (2015). Market basket analysis in insurance industry. Management Science Letters , 5(4), 393-400. </ref>.
+
Indonesia’s motor insurance market has been undergoing dramatic shifts both in demand and supply. GDP is estimated to grow at a steady pace (6%), ushering in larger and more affluent middle class. Both Passenger and Commercial Vehicles have ~10% CAGR. Motorcycles have a positive outlook of 5% CAGR similarly. Prima Facie, this presents a huge market for motor insurance<ref name="kpmg">Indonesia’s Automotive
 +
Industry: Navigating 2014. from https://www.kpmg.com/ID/en/IssuesAndInsights/Documents/Indonesias-Automotive-Industry-Navigating-2014.pdf (Accessed 22 Feb 2015)</ref> .  
 
<br/><br/>
 
<br/><br/>
Kamakura’s<ref name="kamakura">Kamakura, W. (2012). Sequential Market Based Analysis. Springer Science, Business Media, 23, 15-15. doi:DOI 10.1007/s11002-012-9181-6 </ref> work on Market Basket Analysis and Path Analysis is valuable. He demonstrates that Path Analysis can be more useful than Market Basket Analysis in some cases as the sequence of purchase of products is more insightful than the static final basket of goods. Path Analysis can clearly illuminate whether goods are substitutes of each other or complements which provided decision support for cross-selling and bundling of grocery products.
+
However, the government in Indonesia has been sending mixed signals, mostly negative, to the motor insurance industry. Unlike many countries, Third-party insurance is not mandatory and usually clients are ‘forced’ to purchase coverage because it is a condition of the leasing agreement and a majority of Indonesians require financing. These clients seldom renew their policies as it is not mandated. Ernst and Young succinctly summarizes the market condition below:
 
<br/><br/>
 
<br/><br/>
Survival Analysis<ref name="survival-analysis">Introduction to SAS. UCLA: Statistical Consulting Group. from http://www.ats.ucla.edu/stat/sas/notes2/ (accessed November 24, 2007)</ref> models factors or variables that affects the time to an event. Two common methods are used to investigate the survival analysis of a model. First, Nonparametric methods provide simple and quick looks at the survival experience. Second, the Cox Proportional Hazards Regression Model relate the time that passes before some event occurs to one or more covariates that may be associated with that quantity of time. It remains the dominant analysis method for Survival Analysis.
+
“This has caused motor insurers to focus their distribution efforts on building relationships with dealerships and lenders, such as banks and finance companies. It explains why a local insurer with conglomerate links to major dealerships, such as Toyota and Isuzu, holds a dominant place in Indonesia’s motor insurance industry.” <ref name="ey"><Ernst & Young. (n.d.). Motor Insurance : Asia's Growth Engine. Retrieved February 23, 2016, from http://www.ey.com/Publication/vwLUAssets/Motor_Insurance/$FILE/Motor-Insurance.pdf</ref>
 
 
<br/>
 
 
 
===Methodology===
 
Many methodologies are discussed as this group is still in the initial stage of scoping the project. It is estimated that only 1 or 2 objectives will be taken for the project eventually.<br/><br/>
 
 
 
====Objective 1 (Develop a database analysis to formulate a demographic and psychographic profile of customers):====
 
For this objective, we will investigate how the demographics of the customers impact the psychographic/behavioral profile of the customers and investigate the correlation using explorative data analysis and clustering techniques.
 
 
<br/><br/>
 
<br/><br/>
{| class="wikitable" style="width:90%"
+
In 2014, the Finance Ministry of Indonesia (OJK) introduced tariffication which meant that motor insurance companies have to follow a common set of guidelines for pricing of premiums and discounts. According to consulting company Willis International, this meant that motor insurance policies are going to be more standardized and thus competition will be centred largely on pricing <ref name="utomo">Utomo, A. (2014, March 1). NEW ERA IN INDONESIA: TARIFF CHANGES ON PROPERTY AND MOTOR VEHICLE INSURANCE. Retrieved February 23, 2016, from http://www.willis.com/Documents/publications/Services/International/2014/20140410_50261_PUBLICATION_International_Alert_314_FINAL.pdf </ref>.
|-
 
! Some of the demographics we will investigate are: !! Some of the psychographic/behavioral profile we will investigate are:
 
|-
 
|
 
* Age
 
* Gender
 
* Nationality
 
* Profession
 
* Pay
 
* Marital Status
 
* Channel purchased
 
||
 
* Policy purchased
 
* Subsequent Purchase
 
* Profitability of customer
 
* Policy payment default rate
 
* Likelihood of recommending policies to others
 
|}
 
<br/>
 
We can explore demographic segmentation of different nationalities to see if it plays an important role in the customer behaviour. An example is to investigate if Japanese customers are more likely to make subsequent purchase of policies in Tokio Marine than local customers.  
 
 
<br/><br/>
 
<br/><br/>
The investigation of demographic and psychographic profile of the customers will translate into more effective marketing strategies. The company can concentrate its efforts of marketing on demographics of customers that are more profitable and give greater incentive to loyal customers to stay with the firm. Alternatively, the company can also incorporate upselling strategies for customers that are demographically more likely to purchase more expensive policies.
+
Further, in 2012, the government increased downpayment required when leasing motorcycles and cars to 25% and 30% (from 10%)  respectively to prevent a credit risk and property bubble developing. This affected automobile dealers and manufacturers like PT Astra (which distributes Toyota and Honda)  and Suzuki causing their share price to drop by 5% <ref name="bloomberg">Manurung, N., Setiaji,H. (2012, June 8). Indonesia to Push Through Down-Payment Rules Amid Protest. Retrieved February 23, 2016, from
<br/><br/>
+
http://www.bloomberg.com/news/articles/2012-06-07/indonesia-to-push-through-down-payment-rules-amid-protest</ref>. In a dramatic reversal of policy to loosen monetary policy and boost domestic demand, in 2015 the government lowered the downpayment for passenger vehicles and motorcycles by 5% <ref name="kurniati>Kurniati, Y. (2015, June 24). Indonesia Lowers Down Payments for Car, Motorcycle & Property Purchases. Retrieved February 23, 2016, from http://www.indonesia-investments.com/finance/financial-columns/indonesia-lowers-down-payments-for-car-motorcycle-property-purchases/item5676</ref>. On the positive side, the government has slowly moved to a Risk-Based Capital Framework favouring companies with large capital reserves like Tokio Marine. As such in the Indonesian market, consolidation is taking place with smaller domestic players exiting. The top 5 players have about 50% market share in 2010 <ref name="ey"></ref>.
We will also conduct a time series analysis to examine how demographic/psychographic profiles change with time.
 
 
<br/><br/>
 
<br/><br/>
 +
The key issues facing Indonesia’s non-life insurance industry are navigating the regulatory environment, developing new channels of distribution (bancassurance) and managing profitability in this high volume low margin business. The key indicators to pay attention to are Loss Ratio, Average Premiums and Claims Rate and Expenditure. Although benchmarks differ according to geography and degree of competition, it helps to put Tokio Marine Indonesia’s performance in perspective. Generally, the more developed markets have lower profit margins and Multi-National companies tend to dominate. It is against this contextual backdrop that we will proceed with exploration of data.
  
====Objective 2 (Analyse average product holdings per customer): Default rate====
+
<br/>
This objective investigates the length of time a customer continues to hold a product. This investigates the likelihood and the rate at which customers default of periodic policy payments and the rate at which they cancel policies. A high rate of policy cancellation might indicate a flaw with the product that the product management team might need to investigate. The primary techniques to be used for this is Survival Analysis, where Nonparametric methods and Cox Proportional Hazards Regression Model will be used, using the SAS software.
 
<br/><br/>
 
The Nonparametric method used will be a descriptive technique used to provide the rate at which the consumer drops a product. In SAS, a graph of the Kaplan Meier estimate can be used to allows users to see the survival function of the policy changes over time:
 
<br/><br/>
 
[PICTURE GOES HERE]
 
<br/><br/>
 
The Cox Proportional Hazards Regression Model investigates how different variables affect the policy cancellation rate (survival rate) of the policyholders using SAS. The graph below shows the survival function segmented to different age groups:
 
<br/><br/>
 
[PICTURE GOES HERE]
 
<br/><br/>
 
Some potential problems we might face is that most customers hold the policies until maturity. This means that there will be little variance among different groups. We might need to transform the data in order to obtain a more meaningful analysis.
 
<br/><br/>
 
  
====Objective 3 (Determine which customer segments and products are more profitable): Profitability====
+
===Methodology===
In this objective, we will first have to estimate the customer lifetime value of each customer. The customer lifetime value is estimated using some of these variables:
 
* Premium paid
 
* Administration cost
 
* Claim rate
 
* Customer acquisition cost
 
* Retention rate
 
* Policy length
 
* Type of insurance (Whole, Universal, Variable, Term)
 
<br/>
 
Using these variables, we will be able to calculate the margins for each customer. We will then segment the customers to its appropriate segment using clustering techniques. From these segments, we will be able to see the segments that are the most profitable and utilise the appropriate marketing policy.
 
<br/><br/>
 
Tokio Marine should strategically target campaign, provide cheaper deals and provide incentive for renewal to customers that are the most profitable. The company should also try to avoid having policies with customers that eat into margins or charge a higher premium from these customers.
 
<br/><br/>
 
  
====Objective 4 (Which channels are more profitable, direct online or through agents): Channels and profitability====
+
====Exploratory Analysis====
This objective investigates the profitability of products sold by different channels of sales like direct, online or agent. The techniques used will be based on exploratory data analysis and clustering. We will be able to investigate the products each channel typically sells. On a whole, the profitability can be investigated by segmenting the data from different channels and looking at the profitability as a segment.
+
Our group used mainly JMP Pro and Tableau for the initial stages of Exploratory Data Analysis. The Exploratory Analysis was split into motorcycle, vehicle, personal and corporate depending on the variables analysed. Our approach was to look at the general patterns of our data before focusing on the details. JMP is used to do statistical analysis, plotting of treemaps as well as investigation of statistical distribution of the data. Tableau is used for analysis involving a time series as it is more effective in grouping the data by time periods and has more interactivity.
<br/><br/>
 
In order to have an in-depth look on the profitability. we could break down the profitability of different  policies sold by each channel. The difference in profitability might lie in policy cancellation or the different cost of different channels. From these insight, Tokio Marine can order each channels to have a different specialisation of products based on profitability. For example, the online channel can specialise in travel insurance if it is found to be the most profitable channel.
 
<br/><br/>
 
====Objective 5 (Propensity to buy or assess next best offer for customers to enhance effectiveness of marketing campaigns):====
 
Our group will use Affinity Analysis to analyse the customer’s propensity to buy a certain basket of products. After segmenting customers according to demographics, we will conduct an Affinity Analysis to determine which basket of products these segments of customers usually buy. This will then provide guidance to Tokio Marine on marketing a specific group of products more aggressively to a segment of customers or which products to bundle together.
 
 
<br/><br/>
 
<br/><br/>
This can be visualized using a heatmap of co-occurrence as seen in the picture shown below, generated by Tableau:
+
====Time Series Forecasting & Predictive Modelling====
 +
Our group’s predictive model will be using profitability measures like GWP and ClaimPaid, as the dependent variable. We have omitted measures like Commision and Discount as they can be controlled by the company. Our group also aims to create predictive models for different segments like Agents, Brands and Vehicle segment. This allows the company to project which business segment will tend to be profitable.
 
<br/><br/>
 
<br/><br/>
[PICTURE GOES HERE]
+
One technique that our group will be using is the multiple linear regression method. Some of the independent variable were are looking to use in our regression model are Geo_Location, Jap_ind, Corppers, Counter1, Business_ind, Channel, Cover_Type, Area. We will be using measures like t-ratio and VIF to investigate the validity of these independent variables.
 +
Time Series Forecasting & Predictive modelling will only be explored after exploratory analysis is fully completed and the full methodology will be elaborated more in the final report
 
<br/><br/>
 
<br/><br/>
To suggest the next best offer for customers, an Event Sequence Analysis can be used (SAS Enterprise Miner terms it as Path Analysis). This will determine what products customers usually buy given that he has purchased a specific product. This will provide guidance to Tokio Marine on which products to cross-sell and how to more effectively price the products to maximize revenue.
+
====Software Chosen====
 +
The two main software that we used were JMP and Tableau. The reason for using 2 softwares was that each has different strengths. JMP was used primarily for cleaning the data and exploratory data analysis while Tableau was used mainly for visualization and creating interactive dashboards.
 
<br/><br/>
 
<br/><br/>
This can be visualized using a Sankey Diagram as shown below.
+
JMP has several features which makes cleaning data much easier than Tableau. JMP has a large variety of in-built functions to manipulate data tables like ‘update’, ‘join’, ‘split’ et cetera. Unlike Tableau, only the ‘join’ function is in-built and other more complicated  conditional joins have to be written in code. JMP also has the functionality to selectively brush data points on graphs created and it is linked to the data tables as highlighted rows automatically. This is immensely helpful in identifying outliers and subsets within data. Of course Tableau can do this too, however it is more tedious as the user has to manually filter the data on the data table. Furthermore, JMP provides more in-built statistical functions like analyzing simple statistics and other statistical tests like ANOVA.  
 
<br/><br/>
 
<br/><br/>
[PICTURE GOES HERE]
+
Tableau’s strength is that it has more aesthetically pleasing color palettes and more functions for creating storyboards and dashboards which JMP is unable to offer. In creating visualizations, Tableau has the ability to place local and global filters (in dashboards) in the user interface directly. However in JMP, the user has to go back to the data table to manually filter the data. This allows the end user of dashboards to interactively slice and dice the data on the various visualizations intuitively.
 
<br/><br/>
 
<br/><br/>
Both these analytical methods can be done using SAS Enterprise Miner with several defined macros from the community. For further visualization and presentation, the results can be ported over to JMP for its interactivity and range of visualization options.
+
Both JMP and Tableau have the ability to load and export excel files thus working with these 2 programmes concurrently is not a problem.  
 
<br/><br/></div>
 
<br/><br/></div>
  

Latest revision as of 22:52, 28 February 2016

Proposal Interim Review Final


Revised Project Background

Initially, it was mentioned in the proposal that Tokio Marine Life Insurance Singapore was to provide us with their data for analysis. However due to unforeseen circumstances, they were unable to extract and anonymize the data in time. After discussing with our Project Sponsor (Benito Mable), we will be focusing on another dataset which was supplied by TMI as of end January 2016. Objectives were subsequently revised with our Project Sponsor to reflect the different nature of this new dataset.

Revised Motivation

Tokio Marine’s Group Companies (GCs) collect a lot of data required for underwriting products only at the time of sale. Over time, many data points have been captured with little insights derived other than for underwriting purposes. This data is stored on multiple platforms. While some customers have multiple products, it currently is limited in the utilization of data captured to really understand the profile of the customers, what they bought, channel preference, etc.

Tokio Marine’s Asian GCs have been participating in a large-scale regional project, in which the various GCs are undergoing a phase of digital transformation and to stay updated with current technologies. Tokio Marine Asia seeks to convince staff and implement the usage of analytics among the GCs. This project serves as one of the pilot initiatives, with one Asian GC participating - Tokio Marine Insurance Indonesia (TMI). This project aims to use the insights gathered to formulate new marketing initiatives or product ideas.



Data

There will be 2 original datasets provided by Tokio Marine Insurance Indonesia (TMI). The first dataset (“motor_policy30”) contains about 2 million motor insurance policy transaction records, and the second (“motor_claim7_combined”) consists of about 600,000 motor insurance claims transaction records. Both datasets span from 2003 to 2015, and are specific to customers residing in Indonesia only.

The first dataset (“motor_policy30”) has about 152 variables, and the second (“motor_claim7_combined”) has 66, all of which will be included in Appendix A of the Interim Report for reference. Full disclosure of the dataset will only be available to parties which have signed a Nondisclosure Agreement.

The data is of a transactional nature, where it follows a hierarchy such that in the “motor_policy30” dataset, each Policy can have one or more Risk_NO. These in turn correspond to the “motor_claim7_combined” dataset, where each Risk_NO has none or many Claim_NO, where for each there are at least one Transaction_NO. Policies which did not have any claim made under its tenure will not be recorded in the “motor_claim7_combined” dataset.

Example Data:

Example data


Each Risk_NO represents a vehicle insured under the same Policy number, which can be referenced to a particular customer. However, the dataset does not include any customer details.



Claims Process

The multiple Transaction_NO values for each Claim_NO serves to record the transactions corresponding to various steps in the claims-making process.

Claims process.JPG


The columns of ClaimOS, ClaimPaid and ClaimInc emulate the relationship of that of a balance sheet, and can be represented by the equation below:

ClaimPaid + ClaimOS = ClaimInc


Each row of these columns show the transactions applied, but does NOT give the current value of that column. To derive the actual amount of the total ClaimPaid, we can simply sum up the transactions, or use the value in the very last transaction (shown by the cell highlighted in green). This is best illustrated by the example below:

Claims transaction.JPG


Transaction_NO = 1 corresponds to Step 3 in the Claims Process chart, where the ClaimOS is an estimate of the ClaimPaid recorded when Tokio Marine staff have assessed the damage. ClaimInc is subsequently 500,000 as calculated by the equation given. Transaction_NO = 2 corresponds to an adjustment in the estimated figure, to more accurately reflect the actual damage. Transaction_NO = 3 and 4 are the transactions recording the customer receiving the claim amount from Tokio Marine.

Revised Limitations

Limitation Solution
Lack of customer demographic data, as the data is collected only when a claim is made Limit certain analysis to only claims subset data
Lack of other costs like overhead costs which increases cost of each policy Compare only gross loss ratio and not combined ratio with competitors
Complicated business relationships which might skew analysis or have counter-intuitive results Frequent consultations with project sponsor to clarify any surprising and counter-intuitive results




Revised Objectives

  1. Motor Insurance overall profitability and profitability by brand
  2. Motor Insurance and claims trends
  3. Analyse characteristics of Top Agents by Loss Ratio and profitability
  4. Marketing recommendations to improve business performance based on our findings
  5. Time series forecasting of profitability (which will only be discussed more fully in the final report)
  6. Other predictive models (e.g. Multiple Linear Regression), as a bonus


Scope of Work
Due to the sheer number of records in the dataset, we will be restricting our scope to only focus on policies which were underwritten starting 2012 to 2015. Product lines like Shariah/Takaful Islamic motor insurance were discontinued in 2013 so their data will be excluded from the analysis


Research & Methodology

Literature Review

Indonesia’s motor insurance market has been undergoing dramatic shifts both in demand and supply. GDP is estimated to grow at a steady pace (6%), ushering in larger and more affluent middle class. Both Passenger and Commercial Vehicles have ~10% CAGR. Motorcycles have a positive outlook of 5% CAGR similarly. Prima Facie, this presents a huge market for motor insurance[1] .

However, the government in Indonesia has been sending mixed signals, mostly negative, to the motor insurance industry. Unlike many countries, Third-party insurance is not mandatory and usually clients are ‘forced’ to purchase coverage because it is a condition of the leasing agreement and a majority of Indonesians require financing. These clients seldom renew their policies as it is not mandated. Ernst and Young succinctly summarizes the market condition below:

“This has caused motor insurers to focus their distribution efforts on building relationships with dealerships and lenders, such as banks and finance companies. It explains why a local insurer with conglomerate links to major dealerships, such as Toyota and Isuzu, holds a dominant place in Indonesia’s motor insurance industry.” [2]

In 2014, the Finance Ministry of Indonesia (OJK) introduced tariffication which meant that motor insurance companies have to follow a common set of guidelines for pricing of premiums and discounts. According to consulting company Willis International, this meant that motor insurance policies are going to be more standardized and thus competition will be centred largely on pricing [3].

Further, in 2012, the government increased downpayment required when leasing motorcycles and cars to 25% and 30% (from 10%) respectively to prevent a credit risk and property bubble developing. This affected automobile dealers and manufacturers like PT Astra (which distributes Toyota and Honda) and Suzuki causing their share price to drop by 5% [4]. In a dramatic reversal of policy to loosen monetary policy and boost domestic demand, in 2015 the government lowered the downpayment for passenger vehicles and motorcycles by 5% [5]. On the positive side, the government has slowly moved to a Risk-Based Capital Framework favouring companies with large capital reserves like Tokio Marine. As such in the Indonesian market, consolidation is taking place with smaller domestic players exiting. The top 5 players have about 50% market share in 2010 [2].

The key issues facing Indonesia’s non-life insurance industry are navigating the regulatory environment, developing new channels of distribution (bancassurance) and managing profitability in this high volume low margin business. The key indicators to pay attention to are Loss Ratio, Average Premiums and Claims Rate and Expenditure. Although benchmarks differ according to geography and degree of competition, it helps to put Tokio Marine Indonesia’s performance in perspective. Generally, the more developed markets have lower profit margins and Multi-National companies tend to dominate. It is against this contextual backdrop that we will proceed with exploration of data.


Methodology

Exploratory Analysis

Our group used mainly JMP Pro and Tableau for the initial stages of Exploratory Data Analysis. The Exploratory Analysis was split into motorcycle, vehicle, personal and corporate depending on the variables analysed. Our approach was to look at the general patterns of our data before focusing on the details. JMP is used to do statistical analysis, plotting of treemaps as well as investigation of statistical distribution of the data. Tableau is used for analysis involving a time series as it is more effective in grouping the data by time periods and has more interactivity.

Time Series Forecasting & Predictive Modelling

Our group’s predictive model will be using profitability measures like GWP and ClaimPaid, as the dependent variable. We have omitted measures like Commision and Discount as they can be controlled by the company. Our group also aims to create predictive models for different segments like Agents, Brands and Vehicle segment. This allows the company to project which business segment will tend to be profitable.

One technique that our group will be using is the multiple linear regression method. Some of the independent variable were are looking to use in our regression model are Geo_Location, Jap_ind, Corppers, Counter1, Business_ind, Channel, Cover_Type, Area. We will be using measures like t-ratio and VIF to investigate the validity of these independent variables. Time Series Forecasting & Predictive modelling will only be explored after exploratory analysis is fully completed and the full methodology will be elaborated more in the final report

Software Chosen

The two main software that we used were JMP and Tableau. The reason for using 2 softwares was that each has different strengths. JMP was used primarily for cleaning the data and exploratory data analysis while Tableau was used mainly for visualization and creating interactive dashboards.

JMP has several features which makes cleaning data much easier than Tableau. JMP has a large variety of in-built functions to manipulate data tables like ‘update’, ‘join’, ‘split’ et cetera. Unlike Tableau, only the ‘join’ function is in-built and other more complicated conditional joins have to be written in code. JMP also has the functionality to selectively brush data points on graphs created and it is linked to the data tables as highlighted rows automatically. This is immensely helpful in identifying outliers and subsets within data. Of course Tableau can do this too, however it is more tedious as the user has to manually filter the data on the data table. Furthermore, JMP provides more in-built statistical functions like analyzing simple statistics and other statistical tests like ANOVA.

Tableau’s strength is that it has more aesthetically pleasing color palettes and more functions for creating storyboards and dashboards which JMP is unable to offer. In creating visualizations, Tableau has the ability to place local and global filters (in dashboards) in the user interface directly. However in JMP, the user has to go back to the data table to manually filter the data. This allows the end user of dashboards to interactively slice and dice the data on the various visualizations intuitively.

Both JMP and Tableau have the ability to load and export excel files thus working with these 2 programmes concurrently is not a problem.



References

  1. Indonesia’s Automotive Industry: Navigating 2014. from https://www.kpmg.com/ID/en/IssuesAndInsights/Documents/Indonesias-Automotive-Industry-Navigating-2014.pdf (Accessed 22 Feb 2015)
  2. 2.0 2.1 <Ernst & Young. (n.d.). Motor Insurance : Asia's Growth Engine. Retrieved February 23, 2016, from http://www.ey.com/Publication/vwLUAssets/Motor_Insurance/$FILE/Motor-Insurance.pdf
  3. Utomo, A. (2014, March 1). NEW ERA IN INDONESIA: TARIFF CHANGES ON PROPERTY AND MOTOR VEHICLE INSURANCE. Retrieved February 23, 2016, from http://www.willis.com/Documents/publications/Services/International/2014/20140410_50261_PUBLICATION_International_Alert_314_FINAL.pdf
  4. Manurung, N., Setiaji,H. (2012, June 8). Indonesia to Push Through Down-Payment Rules Amid Protest. Retrieved February 23, 2016, from http://www.bloomberg.com/news/articles/2012-06-07/indonesia-to-push-through-down-payment-rules-amid-protest
  5. Kurniati, Y. (2015, June 24). Indonesia Lowers Down Payments for Car, Motorcycle & Property Purchases. Retrieved February 23, 2016, from http://www.indonesia-investments.com/finance/financial-columns/indonesia-lowers-down-payments-for-car-motorcycle-property-purchases/item5676