### Role of Generalised Linear Model in non-life pricing Phase3

- Get link
- Other Apps

Before reading this article, make sure that you read phase1 and phase2. Here are the link:

Phase1: http://www.actuarysense.com/2018/10/role-of-generalised-linear-model-in-non.html

Phase2: http://www.actuarysense.com/2018/11/role-of-generalised-linear-model-in-non.html

Phase1: http://www.actuarysense.com/2018/10/role-of-generalised-linear-model-in-non.html

Phase2: http://www.actuarysense.com/2018/11/role-of-generalised-linear-model-in-non.html

So we know that the purpose of GLM is to find the relationship between mean of the response variable and covariates.

In this Article we are going to talk about Linear Predictors.

In this Article we are going to talk about Linear Predictors.

**: Let’s denote it with, “η” (eta). So, linear predictor is actually a function of covariates. For example, in the normal linear model where function is Y = B**

__Linear Predictor___{0}+ B

_{1}x. So linear predictor will be η = B

_{0}+ B

_{1}x. Always note that linear predictor has to be linear in its parameter. In this case parameters are B

_{0}and B

_{1}.

But still the question is how I came up with B

_{0}+ B_{1}x as a function?
First of all, note that broadly there are two types of Covariates.

1. Variables: It takes the numerical value. For example: age of policyholder, years of experience etc.

2. Factors: It takes the categorical value. For example: Sex of Policyholder, car colour of Policyholder etc.

Let’s see different Scenario:

2. Factors: It takes the categorical value. For example: Sex of Policyholder, car colour of Policyholder etc.

Let’s see different Scenario:

1. If Age is the only covariate that Exists

Linear Predictor:

**η = B**, where we input for X i.e. age of policyholder._{0}+B_{1}x
2. If Age and Sex both are the Covariates (one is factor and other is variable)

Linear Predictor: Age + Sex:

**η = a**, where we input for X i.e. age of policyholder, where i=1 for male and 2 for female._{i}+Bx
3. If Age and Sex are the covariates with interaction between them too.

Liner Predictor: Age + Sex + Age.Sex:

**η = a**so we can see here that with change in “i” value of B also changes_{i}+B_{i}x ,
The reason that this formulation of liner predictor is desirable is its efficiency. In 1

^{st}Case we need to estimate only 2 parameters, in 2nd Case we need to estimate 3 parameter and in last case we need to estimate 4 parameters. So as the covariates keep on increasing, the model will become more complex and we need to estimate more and more parameters.**Follow us on LinkedIn : Actuary Sense**

photo credit: thegeneral.com

So, it simply means that I can estimate as many parameters as data points to make the perfect model? But that will not be the case as it impacts the efficiency of the model. However, we can use that type of model as a benchmark. That type of model is known as “Saturated Model”.

**: It is the model that provides perfect fit to the data. The Saturated model is not useful from a predictive point of view, however it is a good benchmark against which to compare the fit of other models via the scaled deviance.**

__Saturated Model__
So now point is different people came up with their models for pricing of motor insurance, now which model is good and which is not, we can check it using

**.**__Likelihood Ratio Test__
Let’s see the example:

There are 2 models: Model P and Model Q.

Now In Model P, Scaled Deviance will be

**2(l**where l_{S}– l_{P})_{S}represents log likelihood of Saturated model and other one is of Model P. SO now you can relate with topic 1 of this series why we make every Response Variable model into exponential family, so that we can take its log likelihood easily to use it for comparing with other models. 😊
Same thing I can do for Model Q too.

So, now I have scaled Deviance for both models. So we can use Deviance as a measure of the fit of the model. If the data is normally distributed, the Scaled deviance has a Chi-Square Distribution.

**S**=

_{P}- S_{Q }**2(l**then

_{S}– l_{P}) - 2(l_{S}– l_{Q}) = 2(l_{P}– l_{Q})**we can test with Chi-square with 5%(say) significance level.**

Caution: We can only compare two models if one of them is subset of other model. I mean, both models should have same distribution of data and link function and one model is sub model of other model. Through this we can check whether by adding new parameter, my model produce more accurate result or not.

Seems Easy Haan 😊

Thanks and Regards

Actuary Sense

**Follow us on LinkedIn : Actuary Sense**

Reference: Actuarial Education Company 😊

### Popular posts from this blog

### Role of Generalised Linear Model in Non Life Pricing - Phase1

We will cover a series of topics relating to how Non Life Pricing is done through GLM.

But first let's see what is GLM

Let’s take the example of Weight (Y) and Height (X). The aim of linear models is to find the line of best fit through the data points.

Here is your X axis is Height and Y axis is your Weight. Y = B0 + B1x

Line of Best Fit is B0 + B1x where B0 is intercept on Y axis and B1 is the gradient.

Now the question is how that line comes?

Well, line is chosen in such a way to minimize the sum of squared error terms where error terms are distances from data points to straight line, error terms are normally distributed with mean 0 and variance σ2.

2.

We can extend our model to allow for other predictive variables. For example, we can decide that Weight can depend on height and calories consumed per day both. So here we cannot find the line of b…

But first let's see what is GLM

**Generalised Linear Model**Before Jumping on to what is GLM, let’s see what is**Linear models**1.**:***Linear Models*Let’s take the example of Weight (Y) and Height (X). The aim of linear models is to find the line of best fit through the data points.

Here is your X axis is Height and Y axis is your Weight. Y = B0 + B1x

Line of Best Fit is B0 + B1x where B0 is intercept on Y axis and B1 is the gradient.

Now the question is how that line comes?

Well, line is chosen in such a way to minimize the sum of squared error terms where error terms are distances from data points to straight line, error terms are normally distributed with mean 0 and variance σ2.

2.

**:***Multiple Linear Regression*We can extend our model to allow for other predictive variables. For example, we can decide that Weight can depend on height and calories consumed per day both. So here we cannot find the line of b…

### CFM vs UDD vs Balducci

Life Tables: It is a computational tool based on a specific survival model. Our
task is to generate a survival model and our output will be a life table. Life
table is based on Unitary method. lx= Expected no. of lives at age x dx= Expected no. of deaths between exact age x and exact
age x+1 NOTE: We call it expected and not actual because most of the values
will be in decimal too, so how can it be actual.

1.CFM ·It assumes that the force of mortality is constant, i.e. ux+t = ux ·If the force of mortality of a newborn is constant, it means that the expected future lifetime of this life is 1/μ no matter what age he is in. ·Survival Function in this case will decrease exponentially. ·If 0<s<t<1 then t-spx+s = tpx/spx …

*But there is a limitation of Life Tables:*It is defined only for integer ages. If we have to calculate probability of death/survival at any non-integer age we will use 3 assumptions: 1.**CFM=**Constant Force of Mortality 2.**UDD=**Uniform Distribution of death 3.**Balducci assumption**1.CFM ·It assumes that the force of mortality is constant, i.e. ux+t = ux ·If the force of mortality of a newborn is constant, it means that the expected future lifetime of this life is 1/μ no matter what age he is in. ·Survival Function in this case will decrease exponentially. ·If 0<s<t<1 then t-spx+s = tpx/spx …

### MWRR vs TWRR which is better and Why ?

We can decide between various projects that which one
is better and which one is not on the basis of different criteria such as: NPV, IRR, DPP But how can we measure the investment performance?
Well, there are basically three measures of investment performance: 1.Money
Weighted rate of return (MWRR) 2.Time
Weighted rate of return (TWRR) 3.Linked
internal rate of return (LIRR)

It is necessary to measure the performance of a fund which can be a pension fund, funds of an insurance company or funds of an asset management company. It is important for those who are responsible for the investment funds for example: trustees in case of pension fund will monitor how fund is performing i.e. they find out the rate of return of the fund and then compare it with performance of other funds. Before looking at different measures, let’s see some definitions: a.)Income generate by fund: it includes interest payments, dividends received fr…

**Follow us on LinkedIn : Actuary Sense****Follow me on LinkedIn: Kamal Sardana**It is necessary to measure the performance of a fund which can be a pension fund, funds of an insurance company or funds of an asset management company. It is important for those who are responsible for the investment funds for example: trustees in case of pension fund will monitor how fund is performing i.e. they find out the rate of return of the fund and then compare it with performance of other funds. Before looking at different measures, let’s see some definitions: a.)Income generate by fund: it includes interest payments, dividends received fr…

## Comments

## Post a Comment