Introduction

This post documents an end‑to‑end analysis of insurance claims data using R Studio. The main objectives are:

• Measure and analyze the average time to resolution of claims.
• Detect anomalies and potential fraud indicators using statistical and behavioral patterns in the data.

The dataset is synthetic and generic and is designed to simulate typical insurance operations.

Data Dictionary

Below is the structure of the main tables used in the analysis.

Table: Policies

| Column              | Description |
|---------------------|------------|
| `id_policy`         | Unique policy identifier. |
| `insurance_type`    | General type of insurance (e.g., auto, life, health, home). |
| `billing_frequency` | Frequency of premium payments (e.g., monthly, annual, semi‑annual). |
| `issue_date`        | Date when the policy was issued. |
| `start_date`        | Start date of coverage. |
| `end_date`          | End date of coverage. |
| `policy_status`     | Policy status (active, canceled, expired, suspended, renewed). |
| `monthly_premium`   | Monthly premium amount. |
| `annual_premium`    | Annual premium amount. |
| `sum_insured`       | Total insured amount. |
| `deductible`        | Deductible amount per claim. |

Table: Clients

| Column         | Description |
|----------------|------------|
| `id_client`    | Unique client identifier. |
| `gender`       | Gender category (e.g., male, female, not specified). |
| `age_group`    | Age group (e.g., 20‑30, 31‑40, 41‑50, 51‑60, 61‑70, 71‑80). |
| `income_level` | Income level (low, medium, high). |
| `occupations`  | General occupation category (e.g., professional, technical, worker, retired, unemployed, other). |

Table: Claims

| Column                         | Description |
|--------------------------------|------------|
| `id_claim`                     | Unique claim identifier. |
| `id_policy`                    | Foreign key linking to `policies`. |
| `id_client`                    | Foreign key linking to `clients` (optional linkage). |
| `claim_date`                   | Claim occurrence date. |
| `notification_date`            | Date when the claim was reported. |
| `closure_date`                 | Date when the claim was resolved/closed. |
| `claim_type`                   | Claim type (e.g., accident, theft, fire, liability, other). |
| `severity`                     | Severity level (low, medium, high). |
| `claimed_amount`               | Amount claimed. |
| `approved_amount`              | Approved amount after review. |
| `paid_amount`                  | Actual paid amount. |
| `claim_status`                 | Claim status (registered, in process, closed, rejected, under investigation). |
| `resolution_time_days`         | Resolution time in days (calculated as `closure_date` minus `claim_date`). |

Table: Payments

| Column              | Description |
|---------------------|------------|
| `id_payment`        | Unique payment identifier. |
| `id_policy`         | Foreign key linking to `policies`. |
| `payment_date`      | Payment date. |
| `payment_amount`    | Payment amount. |
| `payment_method`    | General payment method (e.g., automatic debit, card, transfer, cash). |
| `payment_status`    | Payment status (paid, pending, overdue, rejected). |
| `days_overdue`      | Days past due (calculated from due date to `payment_date`). |