Monte Carlo Simulation for Energy Risk Management · Monte Carlo Simulation for Energy Risk Management ... •Decision Making Under Uncertainty § Deterministic Analysis § Sensitivity

Monte Carlo Simulation for Energy Risk Management

Scotty Nelson

1

January 15, 2013

Outline of Talk

• Background on Deregulated Power Markets § Regulated vs. Deregulated Power markets§ Market Structure and Participants§ Risk Exposures

• Decision Making Under Uncertainty§ Deterministic Analysis§ Sensitivity Analysis§ Monte Carlo Simulation§ Optimizing the Decision Making Process

•Monte Carlo Simulation§ Model Specification§ Model Estimation§ Model Simulation§ Calibration§ Benchmarking

Analytics for Deregulated Power Markets

•Business questions:§What is my portfolio worth? (valuation)§ How much of my expected dispatch output should I sell

into the forward market? (hedging)§ How much money can I lose? (risk management)§What trades should I enter into so I can maximize my

profits and minimize my risk? (portfolio optimization)

4

Background on Deregulated Power Markets

State of Deregulation

5

Source: Department of Energy

Regulated vs. Deregulated Power Markets

6

Power (MWh)

Load

Generator

Payment ($)

Regulated Setup

ISO

Generator Load

Deregulated Setup

Paym

ent (

$)

Power (M

Wh)

Payment ($)

Pow

er (

MW

h)

Risk Exposure to Power Price MovementsPa

yoff

($)

Power Price ($/MWh)

Generator Load

Payo

ff (

$)

Power Price ($/MWh)

Hedge Optimization

8

9

Decision Making Under Uncertainty

Decision Making Under Uncertainty

• Risk Drivers

§ Deterministic scenario planning models

§ Sensitivity analysis

§ Monte Carlo simulation

• Optimizing the Decision Making Process§ Unconstrained Optimization

§ Constrained Optimization

10

Dispatch Optimization

11

Deterministic Planning Models

• Deterministic planning models§ Pro:

o Simple

§ Con:o How to come up with assumptions?o Are these assumptions realistic?o Doesn’t acknowledge uncertaintyo Can lead to biased decisions

12

Historical versus expected – Henry Hub

13

Historical versus expected – West Hub

14

Historical implied heat rate versus expected implied heat rate

15

Sensitivity Analysis

• Sensitivity analysis

§ Pro:

o Simple

§ Con: o How to create sensitivity scenarios?o Are these scenarios realistic?

• In general the following does not hold, especially for nonlinear functions

16

Monte Carlo Simulation

• Monte Carlo simulation§ Pro:

o Realistic representations of possible states of the world (this could actually happen)o Correlations are maintainedo Can benchmark against actual price distributions

§ Cons:o Complex, slow

17

Optimizing the Decision Process

• Given the prices, we want to optimize a decision process

• Example:§ European Call Option

o Value a call option, value=max(P-K,0) à simple decision rule, if P>K then exercise, otherwise don’t

o Decisions today don’t impact decisions tomorrow

§ Power Plant

o Operational constaints à can’t turn on and off instantlyo How to optimize the decision process, given that decisions today impact possible

decisions tomorrow?o Answer is provided through dynamic programming

18

Dispatch Optimization

19

20

Monte Carlo Simulation

Monte Carlo Framework

- Model Specification

- Specify a model of the fundamental risk drivers

- Model Estimation

- Estimate the unknown parameters of the model

- Simulation

- Simulate the risk drivers

- Calibration

- Use any known information to calibrate the simulations, to match observed real world quantities

- Decision Making

- Optimize the decision process

- Summarize

- Summarize the outcomes (e.g. using probability distributions)

Overview of PowerSimm Processes

22

WX Sim Load Sim

Spot Price Sim

Forward Price Sim

Calibrated Spot Price Data

Dispatch

Portfolio Summarization

Marginal Price of Electricity

$/M

Wh

MW

Supply

Demand

Baseload (Coal) Peakers (CTs)

Marginal price of electricity

Midmerit (CC)

P1

P2

Weather – historical relationships

Weather – modelling – vector autoregression

Weather – simulated temperature – temporal correlations

Weather – simulated temperature – benchmarking

Load – historical relationships

Summer Load Profile Winter Load Profile

Load vs Temperature

Load – modelling – model specification

Load – benchmarking simulations

Load – benchmarking simulations

31

Spot Prices – historical relationships

Spot Prices – modelling

Spot Prices – simulation results

35

Wrapup

Analytics for Deregulated Power Markets

•Business questions:§What is my portfolio worth? (valuation)§ How much of my expected output should I sell into the

forward market? (hedging)§ How much money can I lose? (risk management)§What trades should I enter into so I can maximize my

profits and minimize my risk? (portfolio optimization)

What is My Portfolio Worth?

37

Gross Margin At Risk

Expected Value of Portfolio

How Sensitive is My Portfolio To Prices?

38

Sensitivity of gross margin = $19 million

per $/MWh

Optimal forward sale = ~1500 MW

Questions?

Scotty [email protected]

39