How can demand be determined? The way to determine or project future demand is to use a time series (refer to the article by Gregg Detweiler in the September 1996 First Edition of 1994 - 1998 What Does the Future Have in Store for Wyoming’s Labor Market?). A time series uses what has happened in the past to help predict the future. The data produced are known as projections. These projections show which industries and occupations should be growing or declining in the future. It is useful to know what the future might hold in store for the labor market. It can help individuals make business, educational, and financial decisions based on future projections with some degree of confidence. It also allows for comparison with states in the region and the U.S. (refer to the article by Nancy Brennan in the April 1996 issue of Trends).

There are a number of factors that influence demand (refer to the article by Gayle C. Edlin in the March 1995 issue of Trends). First, projected new openings due to new firms opening or businesses expanding/growing. Second, size of firms in the area. Larger firms usually have greater labor market demand. Third, replacement openings. Not all new openings are due to job creation. Many are replacement openings created when someone leaves a position for any reason. Fourth, seasonal jobs are created at various times in the year. This raises labor market demand at those times of year.

Demand in the labor market is important for all individuals involved in the labor market. Whether as an employer, employee, student, or unemployed, this knowledge can help in planning and decision making. The following articles can help give a better understanding of demand in the labor market.

Trends October 1993, Volume 30 - #10, Brett Judd, pp. 1 - 3
(“Employment Estimates - What do They Tell Us?”)

The diagram depicts the interrelationship between the BLS units in the Research & Planning section. The BLS programs are used to track Wyoming labor market (i.e., labor force, employment, unemployment, and total jobs), with each program counting employment and jobs in a different way. The LAUS program estimates for all people employed and unemployed, while 202 includes only employment covered under Unemployment Insurance (UI), and the CES program is close to the 202 program with some extra inclusions and some exclusions.

ALMIS Training Institute Manual, The LMI Institute, appendix pp. 9 - 10,
(“Labor Turnover Concepts and Measures”)
At any point in time, numerous employers find themselves with an employment level that is less than they currently desire, (i.e., they will experience unfilled job openings or vacancies due to time lags involved in filling job openings that were generated by worker turnover and output expansion). These vacancies can be considered as a measure of unmet-labor demand, (i.e., at the prevailing wage they wish to hire more workers than they currently have on board).

Unmet-labor demand data include the total number of such vacancies, their locations and their industrial and occupational characteristics. Economic policy makers frequently view this information as highly desirable to the planning, design and operation of human-resource programs at the state and local levels. For example, comprehensive job-vacancy data can be used to diagnose existing unemployment problems. And to identify areas of occupational shortage and surplus in state and local labor markets.

Trends February 1997, Volume 34 - #2, Tom Gallagher & Mike Evans, pp. 5 & 8
(“Labor and Population: The First Half of the 1990s and the Balance of the Decade”)
The way an economy values labor influences the probability that labor will be available in a locality. All labor is continuously "bid" by employers in the labor market with some occupations (e.g., retail sales clerks) bid only on a local basis while others (e.g., Computer Scientist) bid internationally. All things being equal (i.e., valued cultural amenities and gasoline prices in Douglas, Wyoming, compared to Lincoln, Nebraska), we can use the available measures of gross population change, employment growth and the value of labor to assess what has happened in the region during the first half of the decade and predict what the second half of the decade may look like for Wyoming.

What Does the Future Have in Store for Wyoming’s Labor Market?, September 1996
First Edition, Gregg Detweiler, p. 4 (“An Overview of Industry Projections”)
A time series is a set of observations obtained by measuring a single variable regularly over a period of time. In a series of employment data, for example, the observations might represent monthly employment levels for several months. A series showing housing starts might consist of weekly housing permits taken over a few years. What each of these examples have in common is that some variable was observed at regular, known intervals over a certain length of time. Thus, the form of the data for a typical time series is a single sequence or list of observations representing measurements taken at regular intervals.

Why might someone collect such data? What kinds of questions could someone be trying to answer? One reason to collect time series data is to try do discover systematic patterns in the series so a mathematical model can be built to explain the past behavior of the series. The discovery of a strong seasonal pattern, for instance, might help explain large fluctuations in the data. Explaining a variable’s past behavior can be interesting and useful, but often one wants to do more than just evaluate the past. One of the most important reasons for doing time series analysis is to forecast future values of the series. The parameters of the model that explained the past values may also predict whether and how much the next few values will increase or decrease. The ability to make such predictions successfully is obviously important to any business or scientific field.

No matter what the primary goal of the time series analysis, the approach basically starts with building a model that will explain the series. The most popular strategy for building a model is the one developed by Box and Jenkins (1976), who defined three major stages of model building: identification, estimation, and diagnostic checking. Although Box and Jenkins originally demonstrated the usefulness of this strategy specifically for ARIMA (Auto Regressive Integrated Moving Average) model building, the general principles can be extended to all model building.

The ARIMA procedure lets you estimate nonseasonal and seasonal univariate ARIMA models. You can include predictor variables in the model to evaluate the effect of some outside event or influence while estimating the coefficients of the ARIMA process. ARIMA produces maximum-likelihood estimates and can process time series with missing observations.

It is often useful to divide your time series into a historical or estimation period and a validation period. One develops a model on the basis of the observations in the historical period and then test it to see how well it works in the validation period. When you are not sure which model to choose, this technique is sometimes more efficient than comparing models based on the entire sample. In our case, the historical period consists of UI covered employment data for all two digit standard industrial codes (SICs) from January 1989 to December 1994. (See 1994 Annual Covered Employment and Wages) The validation period was January 1995 thru December 1995. We used the historical period to find a model to fit or predict the data series in the validation period. Only 15 percent (12) of all the times series that were modeled produced an error more than two percent plus/minus.

For those 12 problematic time series, a different method was used to find a more satisfactory model. The type of modeling used was called Exponential Smoothing. The Exponential Smoothing procedure is best used for short-term forecasting, or what are known as “one-period-ahead” forecasts. When one chooses the right values for its parameters it extracts a lot of useful information from the most recent observation, somewhat less from the next-most-recent, and so on, and usually makes a good forecast. As it moves into the future, however, it quickly runs out of the recent information on which it thrives. However, that is when the analyst’s knowledge about the time series plays an important decision on how to treat each variable.

Trends December 1996, Volume 33 - #12, Lee Saathoff, pp. 1 - 5,
(“Wyoming Occupational Projections: 1994 to 1998")
This article will focus on answering questions about the short-term future of occupations in Wyoming. What occupations are expected to grow? What occupations are projected to decline? Which occupations currently make up large percentages of the Wyoming labor force?

All Occupations as a Whole: Total growth in Wyoming is projected to be 20,341 new jobs over the four-year period from 1994 to 1998, which is 5,130 growth openings per year. This includes 367 occupations that are projected to have positive growth and 56 occupations that are projected to have no growth over the four-year period. In contrast, there are 204 occupations that are projected to decline, for a negative growth of 5,671 jobs over this period or 1,438 annually. Using these figures, we see that Wyoming is projected to have an overall net growth of 14,670 jobs or 3,692 jobs per year. If this level of net growth is realized, Wyoming will increase the level of workers from 215,480 to 230,150 by 1998.

Criteria / OES Subgroup	10000	{20000 & 30000}	40000	50000	60000	70000	{80000 & 90000}	Total
1994 Base Employment	16,336	41,041	22,188	34,245	42,103	2,478	57,089	215,480
1998 Projected Employment	18,283	45,210	24,656	33,023	46,088	2,627	60,263	230,150
1994 % of Total Employment	7.6%	19.0%	10.3%	15.9%	19.5%	1.2%	26.5%	100.0%
1998 % of Total Employment	7.9%	19.6%	10.7%	14.3%	20.0%	1.1%	26.2%	99.8%
Number of Occupations	20	171	21	71	63	18	273	637
% of Occupations	3%	27%	3%	11%	10%	3%	43%	100%
Occupations in Top 50 Growth	4	12	4	4	13	0	13	50
Occupations in Top 200 Growth	11	66	11	14	31	5	62	200
Occupations Declining	7	27	3	41	14	3	109	204
Occupations in Top 50 Declining	3	3	1	21	7	0	15	50
Projected Increase-Net Growth	1,947	4,169	2,468	-1,222	3,985	149	3,174	14,670
Projected Annual Increase	488	1,057	619	-302	997	39	794	3,692

10000 = Managerial & Administrative	60000 = Service
{20000 & 30000} = Professional, Paraprofessional & Technical.	70000 = Agricultural, Forestry, Fishing & Related.
40000 = Sales & Related.	{80000 & 90000} = Production, Construction, Operating, Maintenance & Material Handling
50000 = Clerical & Administrative Support.

Trends April 1996, Volume 33 - #4, Nancy Brennan, pp. 1 - 6, (“Regional Covered Employment and Wage Data: An Economic Indicator for Wyoming”)
Quarterly covered employment and wage data (ES-202 data) in Wyoming are compiled by the Department of Employment, Employment Resources Division, Research & Planning. This data accounts for 85 percent of Wyoming workers. Thus ES-202 data can be used as a strong indicator of the condition of Wyoming's economy. Furthermore, all states must submit their quarterly employment and wage data to the U.S. Department of Labor, Bureau of Labor Statistics. This allows timely evaluation and economic analysis in comparing Wyoming with its surrounding states and the nation.

Trends January 1996, Volume 33 - #1, Carol Toups, pp. 1 - 4,
(“One-Third of Wyoming Employment Found in Small Businesses”)
January 1995 data shows 35.8 percent (72,530) of Wyoming employees working for a small business, with the remaining 64.2 percent (130,237) employed by a large business. Generally, the average monthly employment of the large firm category is higher . . .

Trends March 1995, Volume 32 - #3, Gayle C. Edlin, pp. 1 - 4,
(“Wyoming Full- and Part-Time Demand Occupations”)
The term demand occupations, within the context of this article, refers to occupations for which there was the largest number of job openings between March 1993 and March 1994 (as reported in the Wyoming Wage Survey). Contrary to intuitive logic, the term does not necessarily imply that an increasing number of jobs are created within a particular occupation. The data merely indicate that substantial hiring has occurred, and does not explain why this is so. A large number of job openings within a given occupation may be caused by high employee turnover, for example, rather than by the actual creation of new jobs. Included in the statistical analysis to determine demand occupations were employers who reported in the 1994 Wage Survey (SC&S) that they had workers with less than one year of service in a given occupation (i.e., workers hired or promoted between March 1993 and March 1994). The "Top Ten" full-time demand occupations are summarized in Figure 1; part-time demand occupations are summarized in Figure 2. While some occupations were common to both full- and part-time demand occupations, there were also some which were unique to each category.

Trends July 1994, Volume 31 - #7, Brett Judd, pp. 1 - 3,
(“Seasonal Employment: Benefits and Consequences”)
Although many industries see substantial increases in their employment during the summer months, this article will focus on those industries (eating places, lodging places, business services, and recreation services) which are related to tourism. Not only are these industries increasing during the summer, but the total number of jobs being worked in tourism is increasing each year. The more jobs that are being created is good for the economic welfare of the state, however this expansion may create other economic and labor market issues that need to be addressed.

Trends April 1994, Volume 31 - #4, Nancy Brennan & Valerie Kaminski, pp. 1 - 3,
(“Employment Statistics and Job Training: The Connection”)
Economic activity within the United States, as discussed in last month's Wyoming Labor Force Trends, can be analyzed to help people make educational choices. Labor market information can help to determine which industries are growing or declining.

Table of Contents | Labor Market Information | Wyoming Job Network | Send Us Mail