Wishful thinking validated by selective use of research findings (Psychopath Trump believes that thinking makes it true, but what's the excuse for sane people like us?).
To improve your model, supply and demand would better "identify" supply and demand factors by simultaneous-equations (a supply and demand equation), each with the candidates exogenous factors (determined outside the model) distinct to endogenous demand or to supply (could proxy by in- and out-migration). For example, high-tax, high amenities California, one of the highest demand states ("wouldn't consider living anyplace else") see more outmigration that arrivals (though such "churning" among the highest) because of high costs, taxes, and congestions, from folks who want to live there more; similarly, low-tax Florida, low-public services (with high in-migration has among the highest "flight rates", too, but just not as many). Such a model could then distinguish between the heavy migrations in and out versus states with few moving van traffic, like Maine, W. Virginia, Kansas, and Delaware (which neither attract nor drive out many residents).
One major limiting factor with relying on STATE DATA though (rather than superior Metropolitan Area data) is lack of homogeneity (averages or totals don't represent the bimodality of urban-rural regions); most states have stark divergence between MSA or LMSA job and markets, entrepreneurial opportunities, and amenities versus in medium/small cities and rural; even in (especially in high migration Texas, Calif, Florida, Pennsylvania, New York, Missouri, Illinois, Ohio, Michigan). In addition, be sure to use WEIGHTED REGRESSION (weight by population) due to the 100 to 1 ratios between largest and smallest states, otherwise Wyoming, VT, AK, SD migration way overrepresent results, and CA, TX, FL way underwt'd.
Also, a time series context is important. This century has seen very low rates of interstate migration, marked by notable exceptions such as the mass departure for Louisiana after Katrina, reversal of outmigration states after the Fracking boom in the Dakotas and upper Midwest all the way east to northern PA (N. Dakota went from fastest shrinking among fast pctg. growth).
Amenities (quality of life) that attract and retain people are access to shopping, recreation, major sports teams and cultural events, quality school systems or "choice" options such as charters, public services, urban sprawl, housing affordability, prices, major hub airport, health care, and weather; disamenities include gridlock, pollution, crime, taxes, licenses. States report measures for several of these amenities (average commuting time to work), the other variables may be "proxied" for by common measures (e.g., population of metro area for several). Of course, most people are non-mobile due to constraints from imperfect employment state job markets (e.g., most large universities are in small college towns and often areas with career opportunities for both spouses) or by family proximity (if they feel "place bound" caring for or access to elderly parents or family). Deprived of their favorite amenities, they must attempt to take advantage of nearby inferior substitute amenities instead.
Great to see an analysis of what is deeply felt here in the Democrat controlled tyranny of California.
Wishful thinking validated by selective use of research findings (Psychopath Trump believes that thinking makes it true, but what's the excuse for sane people like us?).
To improve your model, supply and demand would better "identify" supply and demand factors by simultaneous-equations (a supply and demand equation), each with the candidates exogenous factors (determined outside the model) distinct to endogenous demand or to supply (could proxy by in- and out-migration). For example, high-tax, high amenities California, one of the highest demand states ("wouldn't consider living anyplace else") see more outmigration that arrivals (though such "churning" among the highest) because of high costs, taxes, and congestions, from folks who want to live there more; similarly, low-tax Florida, low-public services (with high in-migration has among the highest "flight rates", too, but just not as many). Such a model could then distinguish between the heavy migrations in and out versus states with few moving van traffic, like Maine, W. Virginia, Kansas, and Delaware (which neither attract nor drive out many residents).
One major limiting factor with relying on STATE DATA though (rather than superior Metropolitan Area data) is lack of homogeneity (averages or totals don't represent the bimodality of urban-rural regions); most states have stark divergence between MSA or LMSA job and markets, entrepreneurial opportunities, and amenities versus in medium/small cities and rural; even in (especially in high migration Texas, Calif, Florida, Pennsylvania, New York, Missouri, Illinois, Ohio, Michigan). In addition, be sure to use WEIGHTED REGRESSION (weight by population) due to the 100 to 1 ratios between largest and smallest states, otherwise Wyoming, VT, AK, SD migration way overrepresent results, and CA, TX, FL way underwt'd.
Also, a time series context is important. This century has seen very low rates of interstate migration, marked by notable exceptions such as the mass departure for Louisiana after Katrina, reversal of outmigration states after the Fracking boom in the Dakotas and upper Midwest all the way east to northern PA (N. Dakota went from fastest shrinking among fast pctg. growth).
What high amenities do you believe are part of the California experience?
Amenities (quality of life) that attract and retain people are access to shopping, recreation, major sports teams and cultural events, quality school systems or "choice" options such as charters, public services, urban sprawl, housing affordability, prices, major hub airport, health care, and weather; disamenities include gridlock, pollution, crime, taxes, licenses. States report measures for several of these amenities (average commuting time to work), the other variables may be "proxied" for by common measures (e.g., population of metro area for several). Of course, most people are non-mobile due to constraints from imperfect employment state job markets (e.g., most large universities are in small college towns and often areas with career opportunities for both spouses) or by family proximity (if they feel "place bound" caring for or access to elderly parents or family). Deprived of their favorite amenities, they must attempt to take advantage of nearby inferior substitute amenities instead.