I have served as the principal economist for the ND/MN EB-5 Regional Center since its conception. In addition, to working with the center, I assist clients of other centers with the economic analysis requirements set forth by the USCIS, which include:
To run a successful business, you need to learn about your customers, your competitors and your industry. Market research is the process of analyzing data to help you understand which products and services are in demand, and how to be competitive. Market research can also provide valuable insight to help with risk mitigation, identifying areas in need of improvement, and opportunities.
Biennially I design and implement a survey of stakeholders on behalf of the ND Department of Transportation to evaluate customer satisfaction with the department, providing analysis of key factors contributing to satisfaction and changes over time.
Built a logistic regression model for the University of North Dakota that scores the likelihood students who inquire of our institution will enroll based on geo-demographic characteristics. Read the Paper
I received a grant from the ND Small Business Development Center to demonstrate the use of geodemography for marketing. Read the Report
Forecasting uses historical data (past and present) to infer what to expect in the future. For this reason forecasting is closely related to time-series econometrics techniques. One uses forecasts to identify trends and cycles in the series over time. The accuracy of forecasts will in part be determined on the relevant information set and our ability to closely match the data generating process. Forecasts though are sensitive to structural changes, which create instability in our estimates. Therefore forecasting far into the future can be difficult.
In the past I used time series techniques to build a model of sales from bars and restaurants in Grand Forks to evaluate whether there was any structural change due to passage of legislation prohibiting smoking in workplaces. Unlike previous studies, we found there to be no effect. Read the Final Report
Propensity score matching (PSM) uses statistical methods to create a comparison group that is similar in terms of observed characteristics to the group effected by the policy or program. By matching participants with observationally similar nonparticipants one is able to identify the policy effect. PSM is used when one believes observed characteristics influence participation.
This method was used to examine the effects of orientation programs on academic outcomes of college students. Read the Paper
Difference in Difference methods, compared with propensity score matching (PSM), assume that unobserved differences between individuals/firms influence participation and these differences are time invariant. By using data from prior to and after the program intervention for a control and treated group, one is able to use fixed effects to eliminate the unobserved component and identify the policy effect.
I have used DID to evaluate the effect of the change in participation loan concentration limits on credit union returns. Read the Paper
In addition, I have used the method to examine the effects of changes in student loan limits. Read the Paper
© ND Economist 2015 - Modified from an open source design by Peter Finlan