Modeling Loan Values for Residential Mortgage-Backed Securities
Served as a statistical consultant in the development of dynamic models for residential property valuation across the United States in order to determine whether certain residential mortgage-backed securities (RMBS) were fairly valued. Made use of statistical and econometric techniques including regression modeling, statistical sampling, bootstrapping, and bias adjustment.
Understanding causes of construction delays
Reviewed details of a statistical analysis and a dynamic simulation model, created in order to estimate excess hours spent in construction and design in a delayed power plant project. Work is part of an international arbitration involving the power plant and the construction companies contracted to perform the work. Dr. Salzberg wrote an expert report and testified at an international arbitration on behalf of the owners of the plant.
Determining excess profit
Reviewed and critiqued a regression model to determine excess profit, which was to be used in a sentencing hearing. Work was performed on behalf of the defendants. Dr. Salzberg wrote an expert report setting forth findings using figures rather than formulas to explain statistical concepts.
Product liability data modeling
Reviewed data and “survival analysis” models regarding the number of warranty complaints received for defective windows. Created revised models showing the effects of certain assumptions. Pulled statistical sample to estimate rate of certain types of defects. Dr. Salzberg wrote and expert report and testified in a deposition (on behalf of defendants). The matter is ongoing.
Reviewed data and analysis regarding discrimination claims at the Borgata Hotel in Atlantic City. Wrote an expert report and testified in a deposition on behalf of the plaintiffs.
We have been involved in a number of matters regarding housing discrimination, both on behalf of the plaintiff and on behalf of the defendant. In each case, examined Census data on racial and ethnic makeup of the locality as well as specifics of the proposed housing project in order to determine whether the proposed project would be favorable or disfavorable to minorities.
Ford Windstar back axle defect
In conjunction with our teaming partner, we assisted in analyzing data and performing statistical analyses to determine the possible economic effects of a defect in some models the Ford Windstar’s back axle.
Patent data analysis
Reviewed data related to a patent in order to determine statistical validity of certain claims of the patent. Dr. Salzberg wrote expert reports and testified in a deposition in one of these matters.
Discrimination in highway stops
Assisted in analysis of highway stops of charter buses in a discrimination claim against the state of New Jersey. Our work was on behalf of the state, and helped determine the rate of highway stops for different bus companies, in order to determine if there was a correlation to the race of the companies’ owners.
Analysis of program administration data
On behalf of the Texas Attorney General’s office, reviewed data regarding the state's foster care programs. Wrote an expert report and testified in a deposition related to a matter that claimed insufficient resources were allocated for caseworkers and other activities.
Statistical analysis of unauthorized sales of consumer products
In a matter regarding selling of hair products to unauthorized distributors, analyzed data on thousands of buybacks. On behalf of L’Oreal, performed statistical analysis and determined whether one company was responsible for a disproportionate amount of unauthorized sales. Dr. Salzberg wrote an expert report and testified in a deposition in the matter, which eventually settled.
Analyzing search techniques
Reviewed search techniques, keys, and results for hundreds of thousands of technical searches performed over several years, in order to determine whether the searches were unique, or customized, or whether a number of them were equivalent. Dr. Salzberg wrote and expert report and testified in court on behalf of the company in a sales tax dispute with the state of New York.
Finding "change-points" in stock prices
In a number of matters regarding insider or undisclosed information, used statistical regression modeling to determine the price effects of not disclosing information on the stock price of a company. This form of analysis is frequently termed “event history analysis.”
Pricing analysis in dumping case
On behalf of the United States and working with a teaming partner, analyzed pricing of catfish in order to determine whether it was consisted with market prices or with dumping. Performed statistical analysis using Stata software.
Getting the most out of limited data
Modeled missing data in cable operator surveys in order to determine appropriate share for the Public Broadcasting System. By law, royalties from the retransmission of distant television signals is regulated by a panel of arbitrators at the Library of Congress called the Cable Royalty Tribunal. These royalties are drawn from a pool of funds contributed by local television stations who receive distant signals. Our statistical models made a more objective determination of shares than had otherwise been provided through the expert testimony of economists and survey professionals, and were ultimately accepted by the Cable Royalty Tribunal.
Modeling chances of rare events
Consulted in a litigation matter involving a global engineering and construction company and one of the world's largest energy companies, developed statistical models to estimate the probability of bolts failing along an undersea pipeline, causing a catastrophic leak of oil or gas.
Sampling medical insurance claims files for audit challenges
In a number of matters, a federal agency sought reimbursement for Medicare or Medicaid claims, based on a statistical sample. Retained in some cases by the private client and in others by the government, we pulled statistical samples and/or evaluated the extrapolation results, in order to determine whether the original analysis was appropriate and correct.