|
|
|
-- use this predictive analytics tool to develop & examine projections -- what will change when, where, by how much & with what impact? -- reflecting Census 2020 results, impacts of climate change and pandemic How many single family, owner-occupied housing units will exist in the Houston, TX metro in 2030? This is an example of the use/application of the ProximityOne Modeler software and Situation & Outlook (S&O) database described in this section. The answer to the question, and many others like it, provides stakeholders with projections about the future in their decision-making process -- one more piece of information to help reduce uncertainty. Those who want to examine alternative scenario projections, can use the Modeler to develop and examine those projections. Use Modeler to Develop Estimates & Projections .. goto top Create your own proprietary demogaphic-economic estimates and projections using Modeler or use the Modeler-based demographic-economio datasets developed by ProximityOne. Develop annual/quarterly demographic-economic estimates/projections at the U.S., state, regional, county, tract geographic levels. Integrate your assumptions and business/industry data to examine impact on your organization. 2023 Mid-year Vintage Demographic-Economic Estimates and Projections The 2023 mid-year vintage updates await the Ceneus-sourced model-based county 2022 population estimates by age-gender-race/origin 2022 annual/2022Q4 BLS QCEW data. The ProximityOne S&O estimates and projections developed using Modeler are available in July as datasets, interactively via VDA Web GIS and/or generated generated as alternative scenarios with Modeler. County, Metro, State, National Group: C1 .. all items annually 2020 through 2030 (updates annually) - selected ACS-like DP1-DP4 C2 .. single year of age annually 2020 through 2060 (updates annnually) C3 .. selected items quarterly 2022Q1 - 2024Q1 (updates quarterly) Census Tract: T1 .. tract items annually 2020 through 2030 (updates annually; selected ACS-like DP1-DP4) T2 .. tract single year of age annually 2020 through 2030 (updates annually) All subject matter, geography and time periods synched for additivity and joint, dynamic determination. Examining Trends & Assessing Future Developments .. goto top The economy is always changing. Demographic patterns can change with significant shifts tied to the underlying economy. Demographic-economic changes in one geographic area can often impact the demographic-economic character of an adjacent area. All of these relationship are temporal in nature, changing over time. The behavior of these demographic-economic variables are tied in a statistical-mathematical model that enables the development of projected, as yet unknown, values. To accomplish this, we knit together estimation-projection/modeling software, model specification(s) and historical data. Within the model specification yet other known data are used to form assumptions. More About Modeler The Modeler Winodws-based software operates on a simultaneous equation model. Many demographic or economic estimates and projections are developed without full geographic-economic interaction. Migration in Los Angeles County, CA impacts the demographic change in Maricopa County, AZ, The S&O simultaneous equation model (SEM) has full geographic-demographic-economic integration This model specification provides a more holistic, realistic mathematical-statistical representation of the real world than other more partial models. The S&O SEM is comprised of a set of equations where each specifies the behavior of a jointly dependent variable. The collective sectorized S&O SEM enables the development of approximately 1,000 core subject matter items. Generic County Model ... scroll section .. goto top
Selected equations illustrate structure of the simultaneous equation model.
A generic county level model is used but has some specification differences from county to county.
The generic county model makes use of the population identity equation:
P[i,a,g,r,t] = P[i,a,g,r,t-1] + B[i,a,g,r,t] - D[i,a,g,r,t] + MD[i,a,g,r,t] +MI[i,a,g,r,t] for the ith geographic area; a: age (1, 1, 2, ... 100, 100+); g: gender; r: race/origin; t: year total population for any single area, gender, race/origin, year determined by summing over age (a). Births: B[i,a,g,r,t] determined by fertility rates; where B[i,a,g,r,t] =0 for a>=1 (births only in age 0 cohort) Deaths: D[i,a,g,r,t] determined by mortality rates Domestic Migration: MD[i,a,g,r,t] = f1(MD explanatory variables) International Migration: MI[i,a,g,r,t] = f2(MI explanatory variables) Race/origin categories include, but not limited to: Total population One race alone White alone Black alone American Indian/Alaska Native alone Asian alone Native Hawaiian/Pacific Islander alone Other Race alone, Two or More Races, Hispanic/Latino population (any race) Other identities and cause and effect equations are included in the SEM to determine behavior/values of other items summarized in lists shown below. Examples: Employment: E004[i,a,g,r,t] = f_e004(E004 explanatory variables) School Enrollment: S052[i,g,r,t] = f1_s052(S052 explanatory variables) Educational Attainment, High School Graduate : S061[i,g,r,t] = f_s061(S061 explanatory variables) Households with Income $100,000 to $149,999: E059[i,a,t] = f_e059(E059 explanatory variables) Vacant Housing Units: H003[i,a,t] = f_h003(H003 explanatory variables) Owner Occupied Housing Units: H045[i,a,t] = f_h045(H045 explanatory variables) ProximityOne Demographic-Economic Estimates & Projections .. goto top ProximityOne uses Modeler to develop a range of national scope demographic-economic estimates and projections. These data are available independent of the Modeler package -- as licensed read-to-use datasets or online using the VDA Web GIES. Alternative Scenarios .. goto top A fundamental design view is that even though we seek to develop the best set of estimates and projections, any particular set of estimates and projections reflects just one scenario -- among many possible scenarios. That one scenario makes use of a certain set of known data, models, and algorithms which are implemented/applied through use of Modeler. The models, which purport to explain the cause and effect relationship of the jointly dependent variables (in this case population, components of population, and components of population change), both implicitly and explicitly make use of assumptions. Since assumptions are used to specify quantities or the behavior of things unknown, Modeler should be used to generate estimates and projections for at least three scenarios -- lower (e.g., corresponding to less net in-migration), most likely expected/mid-range, and higher (higher net in-migration). In addition to developing multiple scenario estimates and projections, we often are interested in gaining answers to what-if type questions. Modeler supports this need by enabling the user to make specific assumptions and re-run the development of the estimates and projections. Another consideration in the version or scenario represented is that any particular set of estimates and projections is based on a certain set of historical data. Known event data, such as births and deaths, are updated annually. All variations of previously developed estimates and projections should be replaced with the new, revised estimates and projections based on the latest available data. Estimates versus Projections .. goto top Estimates and projections have geographic, subject matter, and temporal dimensions. Estimates are the data values that are historical though the current time period (2000 through 2023). Projections are the data values corresponding to future time periods (2024 through 2030 .. 2060). A 2023 value might need to be determined by a projection method where historical data only extend to 2022 - an example. The time period spanned by estimates and projections changes with time, annually for Modeler applications. In general, estimates are thought of as known data values typically observed, tabulated, or counted, whereas projections are values conjectured about future data values based on models, assumptions, and other considerations. Updating Estimates and Projections .. goto top All data with a temporal dimension become stale or outdated -- with respect to conveying meaningful information about current circumstances. Several preparation and data development steps were necessary to enable Modeler to perform estimation and projection operations. Some of these steps also require annual or semi-continuous operation/updates. Modeler development and operation might best be viewed as a continuing development/use/dynamic resource, not something updated and used once a year even though a 'new set of estimates and projections' should be developed annually, initially in April-May corresponding to the release of the Census Bureau county level total population estimates. A subsequent adjustment/update should occur in the August-September corresponding to the release of the Census Bureau county level age-race/ethnic-gender estimates. Custom Estimates and Projections for Specialized Geographies & Universes The Situation & Outlook estimates and projections are developed for a standardized set of geographies. These areas include the U.S., states, metropolitan areas, counties, school districts, cities and census tracts. In the case of school districts, estimates and projections are developed for the school district community, but not public versus private school enrollment, enrollment by grade nor enrollment by attendance zone. School enrollment estimates and projections are facilitated by geocoding students enrolled by residence and summarizing these data at the census block, block group and census tract levels. The set of enrollment estimates and projections can be developed through use of a set of additional equations added to the simultaneous equation model. Hard-to-Estimate Populations .. goto top ProximityOne specializes in hard-to-estimate populations. We have developed estimates and analyses related to Native American populations ranging from Idaho to Oklahoma. We have developed estimates and projections of Native Hawaiians to meet educational planning needs. We have developed estimates and projections college-age population by type of enrollment by census tract in Georgia to help assess underserved college-age population by the Georgia state university system. Support Using these Resources .. goto top Learn more about accessing and using demographic-economic data and related analytical tools. Join us in a Data Analytics Lab session. There is no fee for these one-hour Web sessions. Each informal session is focused on a specific topic. The open structure also provides for Q&A and discussion of application issues of interest to participants. ProximityOne User Group .. goto top Join the ProximityOne User Group to keep up-to-date with new developments relating to metros and component geography decision-making information resources. Receive updates and access to tools and resources available only to members. Use this form to join the User Group. Additional Information ProximityOne develops geodemographic-economic data and analytical tools and helps organizations knit together and use diverse data in a decision-making and analytical framework. We develop custom demographic/economic estimates and projections, develop geographic and geocoded address files, and assist with impact and geospatial analyses. Wide-ranging organizations use our tools (software, data, methodologies) to analyze their own data integrated with other data. Follow ProximityOne on Twitter at www.twitter.com/proximityone. Contact us (888-364-7656) with questions about data covered in this section or to discuss custom estimates, projections or analyses for your areas of interest. |
|
|
