Demand Equation For Natural Gas Usage

Demand Equation for Natural Gas Usage

Introduction

Clearwater Enterprises LLC is a natural gas marketing and consulting company based in Oklahoma City. In addition to marketing natural gas for producer clients throughout the state, Clearwater Enterprises provides retail gas sales services to more than 1,200 end users and consumer clients. Clearwater Enterprise's goal is to provide competitively priced reliable natural gas to its clients and advising them on the current natural gas climate. In order to provide competitively priced natural gas, Clearwater Enterprise's from time to time, stores natural gas for its clients and takes pricing risk in a volatile gas market. Being long or short gas in the company's storage account can cost money and affects the bottom line. Given the close relationship between daily high and low temperatures and natural gas usage, demand estimation or forecasting is very vital for Clearwater, sand is critical in avoiding such additional fees.

This term-paper is prepared for how short-term weather forecast is affecting natural gas usage for Clearwater and its clients and intended as a reference for the company when estimating short-term natural gas usage.

Demand Curve Estimation

Demand curve estimation is sometimes relatively simple, especially in the case of stable short-run demand relations. The best technique for estimating the demand curve is the method that provides a necessary level of accuracy at minimum cost. In many instances, simple demand curve estimation techniques are the most effective and cost efficient approach. Simple methods are commonly used to estimate demand curve. If a manufacturer has a substantial backlog of purchase orders, the pace of future sales can sometimes be estimated precisely. But the demand estimation still involves errors, even when a large and growing backlog of customer orders is evident. The dynamic nature of demand relations makes it tough to accurately estimate demand, and even tougher to determine the effect on demand of modest changes in prices, advertising, credit terms and so on. This is especially true for cyclical goods such as household appliances, machine tools and raw materials. If we assume natural gas as a commodity is a non-cyclical good and that the demand is relatively unaffected by changing income, some of the assumptions stated above would not affect demand for natural gas consumption. There are, however several different factors that may affect natural gas demand, which include weather, household formation, commercial employment, natural gas prices relative to competing fuel prices and industrial output. These are all important factors in the short-term determination of natural gas demand. For our case purposes, we will be using only daily high and low temperatures as independent variables and apply regression analysis to data to determine a relationship between temperature and daily usage

Regression Analysis

Regression analysis is a powerful statistical technique that describes the way in which one important economic variable is related to one or more other economic variables. Although there are clear limitations to the technique, regression analysis is often used to provide successful managers with valuable insight concerning a variety of significant economic relations. A statistical relation exists between two economic variables if the average of one is related to another. However it is impossible to predict with absolute certainty the value of one based on the value of another. The exact relation between two variables is not known for certainty and must be estimated. The most common means for doing so is to gather and analyze historical data. We will be using Clearwater's first quarter 2007 actual daily natural gas data to formulate a demand equation to be used in short term consumption forecasts.

The first step in regression analysis is to specify variables to be included in the regression model. Historical daily natural gas usage measured in thousand cubic feet (Mcf) will be our dependent variable when specifying our demand function. Our independent variables will be daily high and low temperatures. The second step in regression analysis is to obtain reliable data. Once variables have been specified and the data have been gathered, the functional form of the regression equation must be determined. This form reflects the way in which independent variables are assumed to affect the dependent variable. The most common specification is a linear model such as the following demand function; Q = B0 + B1X.

The most popular regression technique is using ordinary least squares to estimate the coefficients for linear equations. The method of least squares estimates, or fits the regression line that minimizes the sum of the squared deviations between the best-fitting line and the set of original data points. The technique is based on the minimization of squared deviations to avoid the problem of having positive and negative deviations cancel one another out.

If a regression analysis involves only one dependent and one independent variable, such a regression equation is called simple regression. In order for us to predict short-term natural gas demand, we will be using multiple regression models which include two or more independent variables. Table below shows 1st quarter's natural gas usage and daily temperatures for Clearwater.

Date Daily Usage (Mcf) High Temp. Low Temp

1/1/2007 28,447 49 28

1/2/2007 32,915 51 24

1/3/2007 29,847 59 29

1/4/2007 27,914 51 44

1/5/2007 27,061 58 44

1/6/2007 25,468 48 33

1/7/2007 31,160 50 29

1/8/2007 31,781 60 23

1/9/2007 30,571 55 31

1/10/2007 27,732 64 34

1/11/2007 24,320 71 53

1/12/2007 38,149 58 22

1/13/2007 38,464 24 21

1/14/2007 42,889 26 22

1/15/2007 45,459 25 16

1/16/2007 45,513 26 18

1/17/2007 43,449 32 22

1/18/2007 39,699 39 31

1/19/2007 36,175 39 27

1/20/2007 35,171 34 31