Taylor and McGuire Flashcards

1
Q

Mean and variance of exponential dispersion family (EDF) of distributions

A

mean = mu

variance = dispersion parameter * variance function

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Variance function for the Tweedie sub-family of EDFs

A

variance = mu ^ p
And p between 0 and 1 (inclusive)

in words: variance is proportional to a power of the mean

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Relationship between p for the Tweedie distribution and tail heaviness

A

tail heaviness increases as p increases

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Mean and variance of a Tweedie distribution

A

mean = mu = [ ( 1 - p ) * theta ] ^ ( 1 / ( 1 - p ) )

where theta = location parameter

variance = dispersion parameter * mu ^ p

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

General GLM format (in matrix notation)

Taylor and McGuire

A

link function = transposed covariate matrix * beta matrix

where betas are the linear response variables, and the link function transforms the mean of each observation into a linear function of the parameters (betas)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Conditions for the structure of a GLM (3)

Taylor and McGuire

A
  1. each observation is a member of the EDF
  2. h(mu_i) = x^T * B
  3. observations are stochastically independent
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Underlying assumptions of a standard linear regression (3)

A
  1. errors are normally distributed
  2. errors have constant variance
  3. linear relationship between X and Y
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Difference b/w weighted linear regression and standard linear regression

A

weighted linear regression recognizes errors might have unequal variances

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Main differences between a GLM and a linear regression (2)

A
  1. non-linear relationship between X and Y
  2. non-normal error term
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Common estimation method for GLM parameters

A

MLE

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Requirements for selection of a GLM and purpose of each (4)

A

selection of:

  1. cumulant function (controls the shape of the distribution)
  2. index, p (controls relationship b/w mean and variance in an EDF)
  3. covariates (x’s = explanatory variables)
  4. link function (controls relationship b/w mean and covariates)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Measure of model goodness of fit

Taylor and McGuire

A

deviance

> > smaller = better

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Deviance formula (unscaled)

A

deviance = 2 * sum ( log-likelihood (perfect model ) - log-likelihood ( actual model ) )

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Scale parameter calculated from deviance and corresponding distribution
(Taylor and McGuire)

A

scale parameter = deviance / ( n - p )

> > Chi-square distribution w/ ( n - p ) df

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Standardized Pearson Residuals (Taylor and McGuire)

A

= raw residual / std. dev. ( observation )

  • unbiased and homoscedastic
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Problem with standardized Pearson residuals

Taylor and McGuire

A

does not remove skewness from the data

we prefer residuals that are approximately normally distributed

17
Q

Best residual to use for model assessment and why

Taylor and McGuire

A

deviance residuals

why: corrects any non-normality in the data

18
Q

Deviance residual

A

Rd_i = sqrt(d_i / phi^hat) * sign(Y_i - Y_i^hat)

19
Q

Types of stochastic models (4)

Taylor and McGuire

A
  1. non-parametric Mack model
  2. parametric Mack models
  3. cross-classified models
  4. GLM representations of CL models
20
Q

Results of non-parametric Mack model (2)

A
  1. estimators of CL age-to-age factors are MVUE’s among estimators that are unbiased linear combinations
  2. unbiased CL reserve estimates
21
Q

Special cases of parametric Mack models (2)

A
  1. ODP Mack model
  2. Tweedie Mack model

these remove Variance assumption from Mack, variance confined to the variance of an EDF distribution

22
Q

Assumption required to turn a non-parametric Mack model into a parametric one

A

require that incremental observations (given claims to date) come from the EDF distributions

23
Q

Theorem 3.1 from Taylor and McGuire (3 MVUE results for parametric models)

A

under EDF and general Mack assumptions:

  1. MLEs are the unbiased CL estimators
  2. for ODP Mack and column dispersion parameters, CL estimators are MVUEs
  3. cumulative loss and reserve estimates are also MVUEs
24
Q

Reason Taylor and McGuire’s parametric MVUE results are stronger than regular Mack results

A

minimum variance of all unbiased estimators, not just linear combinations

25
EDF cross-classified model assumptions (2)
1. stochastic independence of response variable 2. explicit row and column parameters, where column parameters sum to 1
26
Theorem 3.2 from Taylor and McGuire (EDF and ODP cross-classified results)
under the EDF cross-classified assumptions, restricted to an ODP distribution with constant dispersion parameter, the MLE fitted values and forecasts are the same as the usual CL method
27
Theorem 3.3 from Taylor and McGuire (MVUE for cross-classified models)
if theorem 3.2 applies and the fitted and forecasted values are corrected for bias, then they are MVUEs
28
Alpha and beta parameter calculations for non-GLM version of ODP cross-classified model, order of calculations, and forecasted incremental losses
order: alpha increasing, beta decreasing alpha ( 1 ) = latest cumulative loss all other alphas = latest cumulative loss / ( 1 - sum of already calculated betas ) beta = sum ( incremental losses in column ) / sum (already calculated alphas ) forecasted incremental losses = alpha * beta for given row and column
29
Difference b/w ODP Mack model and ODP cross-classified model under GLM representations of CL models
ODP Mack models link ratios ODP cross-classified models incremental losses
30
Matrix notation for GLM representation of ODP Mack model
Y = X * beta where Y = individual predicted age-to-age factors (all) X = identity matrix beta = volume weighted age-to-age factors
31
Matrix notation for GLM representation of ODP cross-classified model
Y = X * beta where Y = estimated incremental losses (all) X = identity matrix beta = ln of all alpha and beta parameters (single column in order)
32
Problem with GLM representation of ODP cross-classified model and consequence
can lead to parameter redundancy, need to alias a parameter (GLM software will do this automatically) >> if parameters are aliased, estimates will not match the non-GLM version (rescale alpha and beta parameters)
33
Forecast design matrix (GLM ODP cross-classified model)
same as regular GLM representation but only shows estimates of future incremental losses