Probability Provides a Measurable and Quantifiable Indication

Probability provides a measurable and quantifiable indication of how likely a particular outcome is for an event or experiment. It is expressed as a numeric value between zero and one (inclusive), where an impossible event has a probability of zero and a certain event has a probability of one. Probability is conventionally written using the symbol P, and may be expressed as a percentage by multiplying P. By 100.

The problem at hand related to 1995 gold prices can be set up and solved in the following way, with µ (mu) representing the mean, ? (sigma) the standard deviation, X the target value, and x1 and x2 defining the target value range.

µ = $383; = $12; x1=$394; x2=$399

The probability that the client's gold will be sold the next day is given by:

P (x1

Whereas it is possible to compute this result manually using z-values and standard statistical tables, automated calculator tools with simple user interfaces are available online for this purpose. For example, the calculation above was enabled using the online tool provided by Dendane (2011).

2. Statistical sampling is vitally important for both academic pursuits and for the conduct of business. Generally, sampling involves selecting a small and manageable number of individuals from a population to enable predicting the properties of the population as a whole. This is known as inferential statistics, and is key to business decision making. The primary justifications for using sampling in business instead of a comprehensive census include increased cost savings and timeliness for decision making due to reduced data collection effort, increased accuracy through reduction of errors in data collection, and increased informational value through advanced data processing upon the smaller sample set. In addition, in cases where studying a sample could result in the alteration or destruction of the item under observation, losses due to sample destruction can be reduced to acceptable levels for the application at hand. Randomized sampling in particular is preferable for the purpose of statistical analysis, as it helps to ensure that each member of a population has an equal chance of being selected for the sample, without being subjected to the possibility of a biased selection process that could skew the results. In certain situations, sampling of a population may be unnecessary or ineffective. For example, if a single instance of a condition is sufficient to either prove or disprove an assertion, then examining a probability sample may not be needed to establish the result. Alternately, in the case where subjective insights or exploratory results are all that is desired with no requirement to represent the overall population, statistically valid random sampling may be optional.

3. Regression analysis is an important topic of study in a business education curriculum, as it is a commonly used method for forecasting business parameters as well as for analyzing trends in business data. Spreadsheet software such as Excel by Microsoft are typically used as enabling software for regression analysis, and require building a base of competency to use effectively in business. Regression models are used in sales, financials and technology applications. Regression analysis involving a single independent variable is known as simple linear regression. This technique is commonly used for business activities such as generating sales or budgetary forecasts on a quarterly or annual basis. In simple linear regression, the core statistical concepts involved include variables, covariance, correlation coefficients and the linear regression equation. For situations involving two or more independent variables, a more sophisticated method may be needed. In such cases, a technique known as multiple linear regression may be employed to provide insights into the relationship between a dependent (criterion) variable and one or more independent (predictor) variables, which can individually or jointly affect the outcome. An example of a business use of multiple regression is real estate valuation, where several physical parameters of a property (predictors) may used to help predict the selling price range of the property (criterion).

4. In this section, we will examine the role of statistical concepts in the context of the food service industry. Specifically, we will discuss the applicability of frequency distributions and sampling, as well as how probability and measures of central tendency have practical uses in food services. Statistical methods are routinely used in the sales and marketing of food services, and also play a critical role in ensuring quality and safety of foods and related services by helping to reduce process and output variability.

Frequency distributions are typically used to predict customer spending patterns for foods and related services in different target environments. Statistical sampling analysis is then used to confirm and adjust these forecasts to drive decisions about marketing, supply, operations and logistics. A practical example of the use of frequency distribution and sampling is the analysis of the typical amount spent by families at amusement parks and their spending patterns based on the type of food or time of day.