Economics – Demand Estimation Sample Essay

What Is Demand Estimation?
When running a little concern. it is of import to hold an thought of what you should anticipate in the manner of gross revenues. To gauge how many gross revenues a company will do. demand appraisal is a procedure that is normally used. With demand appraisal. a company can estimate how much to bring forth and do other of import determinations.

Demand appraisal is a procedure that involves coming up with an estimation of the sum of demand for a merchandise or service. The estimation of demand is typically confined to a peculiar period of clip. such as a month. one-fourth or twelvemonth. While this is decidedly non a manner to foretell the hereafter for your concern. it can be used to come up with reasonably accurate estimations if the premises made are right. Methods of Demand Estimation:

We will write a custom essay sample on
Economics – Demand Estimation Sample Essay
or any similar topic only for you
Order now

There are a assortment of ways that can be used to gauge demand. each of which has certain advantages and disadvantages. They are divided into Qualitative and Quantitative Methods. Qualitative Methods:

Qualitative methods consists of following points. Consumer studies:
Firms can obtain information sing their demand maps by utilizing interviews and questionnaires. inquiring inquiries about purchasing wonts. motivations and purposes. These can be speedy on-the-street interviews. or in-depth 1s. They might inquire. for illustration. how much more petrol respondents would purchase if its monetary value were reduced by Rs. 15 per litre. or which trade name of several possibilities they prefer. These methods have certain drawbacks. Market experiments:

As with consumer studies these can be performed in many ways. Laboratory experiments or consumer clinics seek to prove consumer reactionsto alterations constants in the demand map in a controlled environment. Consumers are usually given little sums of money and allowed to take how to pass this on different goods at monetary values that are varied by the research worker. However. such experiments have to be set up really carefully to obtain valid andreliable consequences ; the cognition of being in an unreal environment can impact consumer behavior other types of market survey involve utilizing existent markets in different geographic locations and changing the governable factors impacting demand.

Quantitative methods:

Statistical methods

While the above methods are utile. they frequently do non supply direction with the sort of elaborate information necessary to gauge a utile demand map. and thereby prove the relevant hypotheses and do prognosiss. Statisticaltechniques. particularly regression analysis. supply the most powerful agencies of gauging demand maps. Regression techniques do have assorted restrictions: 1 ) They require a batch of informations in order to be performed.

2 ) They necessitate a big sum of calculation.
3 ) They suffer from a figure of proficient jobs.
In malice of these restrictions. arrested development techniques have become the most popular method of demand appraisal. since the widespread handiness of powerful desktop Personal computers and package bundles have made at least the first two jobs easy to get the better of.

Model specification:
There are two major facets of this phase. In order to understand this we must foremost separate a statistical relationship from a deterministic relationship. The latter are relationships known with certainty. for illustration the relationship among gross. monetary value and measure: R=P*Q ; if P and Q are known R can be determined precisely. Statistical relationships are much more common in economic sciences and affect an component of uncertainness. The deterministic relationship is considered foremost. Mathematical theoretical accounts:

It is assumed to get down with that the relationship is deterministic. With a simple demand curve the relationship would hence be Q=f ( P )
If we are besides interested in how gross revenues are affected by the past monetary value. the theoretical account might in general become Qt=f ( Pt. Pt-1 ) where Qt represents gross revenues in one month. Pt represents monetary value in the same month andPt-1 represents monetary value in the old month. This last variable. affecting graph values in a old clip period. is known as a lagged variable. Other variables could besides be included on the right manus side if economic theory or old empirical surveies indicated that they might be of import. The determination sing which variables to include is a hard one. Theory frequently tells us that certain variables. like monetary value. publicity and income. should impact gross revenues. but before we collect the informations and analyse the consequences we do non cognize for certain which variables are relevant ; in fact. even after analysing the information we do non cognize for certain which variables are of import because we are gauging a relationship from sample informations. and therefore we can merely do decisions in probabilistic footings. Therefore there is ever a gray country if a priori economic theory struggle.

Statistical theoretical accounts:
In pattern we can really seldom stipulate an economic relationship precisely. Models by their nature affect simplifications ; in the demand state of affairs we can non trust to include all the relevant variables on the right manus side of the equation. for a figure of grounds 1. We may non cognize from a theoretical point of view what variables are relevant in impacting the demand for a peculiar merchandise. 2. The information may non be available. or impossible to obtain. An illustration might be the selling outgos of rival houses. 3. It may be excessively dearly-won to obtain the relevant information. For illustration. it might be possible to obtain information associating to the income of clients. but it would take excessively much clip ( and may non be dependable )

If we simplify the relationship to merely two variables the spread graph given shows that the relationship is far from perfect ; in a perfect relationship the points would precisely suit a consecutive line. or some other regular curve. We hence have to stipulate the relationship in statistical footings. utilizing a residuary term to let for the influence of omitted variables. This is shown for the additive signifier as follows Qi=a +bPi +di where di represents a residuary term. Thus. even if P is known. we can non foretell Q with complete truth because we do non cognize for any observation what the size or way of the remainder will be. Data aggregation

Statistical methods place a large demand on informations ; hence. the aggregation of informations is important in pattern. This phase is frequently ignored in the sorts of jobs with which pupils are often faced. where they are already presented with informations in a useable signifier ; this phase of the analysis is besides normally discussed in more item in market research classs. Three facets are discussed here: types of informations. beginnings of informations and presentation of informations. Types of informations

There are two chief types of informations that houses can roll up.
a. Time series Data
This refers to data on a individual entity at different periods of clip. These informations can be collected yearly. quarterly. monthly or at any regular interval. Thus gross revenues of house A in the period 1994–99 would affect clip series informations. Such informations may be quantitative. intending that they are measured numerically on an ordinal or central ; illustrations are gross revenues. monetary values and income. Alternatively. informations may be qualitative. intending that they are nominal. or expressed in classs ; illustrations are male/female. married/single/widowed/divorced. east/west. The intervention of such variables. frequently called dummy variables. is considered. under extensions of the theoretical account. b. Cross-section information

This refers to data on different entities at a individual period of clip. In managerial economic sciences these entities are usually houses. therefore gross revenues of houses A-F in 1999 would affect cross-section informations. Sometimes the entities are persons. industries or countries. The different types of informations have certain advantages and disadvantages to the research worker. In pattern the research worker may hold small pick. because merely one type of informations may be available or executable to utilize. Sometimes the two types of informations can be pooled. that is combined together. For illustration. a survey of six houses over six clip periods would give 36 observations ; such informations allow more observations. which is an advantage in analysis. However. pooling information has to be done with attention to avoid jobs of reading. Beginnings of informations

In pattern we should seek to roll up informations associating to all the variables that we thin thousand might impact gross revenues. on either a time-series or cross-section footing. harmonizing to how we have specified the theoretical account. Later. after the statistical analysis. some of these variables may be omitted. There are many beginnings of informations available. but in general the followers are the most of import in demand appraisal. and so in most of managerial economic sciences. 1 ) Records of houses.

Gross saless. selling and accounting sections maintain records of many of the cardinal variables of involvement. Such informations are usually up to day of the month. 2 ) Commercial and private bureaus. These include consulting houses. market research houses and Bankss. In add-on. a house may desire to committee one of these bureaus to transport out a peculiar survey. but it would hold to see the cost involved compared with utilizing freely available informations. 3 ) Official beginnings.

These include authorities sections and bureaus. and international bureaus like the EU. OECD. WTO and the assorted UN bureaus. Such informations tend to be more macroeconomic in nature. although there are besides many industry surveies. The information may besides be slightly out of day of the month. since it takes clip to roll up. collate and print it. sometimes every bit long as a twosome of old ages. Much of the above information is now available on the Internet. peculiarly those from the 3rd beginning and some of those from the 2nd. It is of import to appreciate that the usage of any of the above beginnings. whether published on the Internet or non. involves abstraction. This means utilizing informations that have been collected by person else ; such informations are often referred to as secondary informations. Although it is evidently easier and cheaper to utilize such informations. there are restrictions of which the research worker has to be cognizant. The informations have been collected for a different intent from that of the current probe and the research worker does non cognize the conditions under which the informations were collected. The definitions used may be different from those now desired. For illustration. the monetary value variable measured and recorded in a firm’s records may be the quoted monetary value. non the existent monetary value leting for any price reductions. Clearly it is the 2nd step that is of import in demand appraisal. but the research worker may non be cognizant of the original definition used.

Presentation of informations
a ) Tables: The most basic method of showing demand informations is in the signifier of a tabular array. To get down with. we will take a two-variable survey. affecting merely measure ( gross revenues ) and current monetary value. to simplify the analysis. In world this is merely justified if either:1 ) No other variables affect gross revenues ( extremely unlikely ) . or2 ) Other variables do affect gross revenues but remain changeless ( still reasonably improbable ) . The chief advantage of restricting the survey to two variables is that such relationships can easy be shown diagrammatically. See the illustration in Table 2. associating to a cross-section survey of seven houses. The ground for entering the monetary value variable in the last column. after graph column show regular increases of one unit ; although one is improbable to happen such regularity in pattern ; it simplifies the numerical analysis and allows easier penetrations every bit far as statistical illation is concerned. Table 2

B ) Graph: In order to analyze the relationship more closely the following measure is to pull a graph. There are two chief rules involved here: ?
Gross saless ( Q ) should be measured on the perpendicular axis as the dependant variable ; this is contrary to most price–quantity graphs. ? Scales should be selected so as to hold the informations spread over the whole graph ; this involves looking at both the highest and lowest values in the information. Scales should non therefore automatically get down at nothing. The consequence is a spread graph. as shown in Figure 1 ; no effort is made to fall in together in any manner. We can see several things from this graph.

There is by and large an opposite relationship between the variables.
The relationship is non a perfect one ; the points do non lie precisely on a consecutive line or hyperbola. This is because of the skip of other variables impacting gross revenues. significance that the premise made earlier sing these variables ( that they did non impact gross revenues or remained changeless ) was non wholly justified

OLS ( Ordinary Least Squares ) Method for Arrested development:

The method of least squares means happening the line that minimizes the amount of the squares of the differences between the ascertained values of the dependant variable and the fitted values from the line. To set it mathematically. we need to happen an equation Y=aX+b which minimizes the amount of squared divergences ? ( Y-Y ) 2. where Y is the estimated value of the dependent variable as per the fitted curve. The technique for work outing for the values of A and B is to utilize partial distinction with regard to both a and b. set both looks equal to zero to minimise them. and work out them at the same time. The resulting solutions are as follows

More specifically the correlativity coefficient ( R ) measures the grade of additive association between variables. It should be noted that correlativity says nil about causing. The causing between the variables could be reversed indirection. or it could move in both waies in a round mode. For illustration. high gross revenues could take to economic systems of graduated table in production. enabling houses to cut down their monetary value. An alternate account of correlativity between variables is that there may be no causing at all between two variables ; they may both be influenced by a 3rd variable.

The coefficient of finding
The job with the correlativity coefficient is that it does non hold a precise quantitative reading. A better step of goodness of tantrum is the coefficient of finding. which is given by the square of the correlativity coefficient. and is normally denoted as R2 This does hold a precise quantitative reading and it measures the proportion of the entire fluctuation in the dependant variable that is explained by the relationship with the independent variable. In order to understand this step more to the full it is necessary to analyze the statistical construct of fluctuation and the constituents of explained and unexplained fluctuation. This is best done with the assistance of a graph ( see figure below )

Demand Estimation and Fore casting. In statistical footings. fluctuation refers to the squared divergences. Thus the entire fluctuation in Y is the amount of squared divergences from the mean of Y. or the entire amount of squares ( TSS ) . However. for each X. Entire Deviation or TD. can be partitioned into two constituents. explained divergence ( ED ) and unexplained divergence ( UD ) . The first constituent is explained by the arrested development line. in other words the relationship with X.


Hi there, would you like to get such a paper? How about receiving a customized one? Check it out