A Missing Variable Imputation Methodology with an Empirical Application

Kyureghian, Gayaneh; Capps, Oral; Nayga, Rodolfo M.

doi:10.1108/s0731-9053(2011)000027a015

Cited by 19 publications

(9 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kyureghian et al . [] outline and test six methods used to impute missing data using two data sets that feature “missingness” identical to that encountered in our data (that is, missing household‐level price information for goods not purchased). The authors concluded that, on average, the conditional mean—a regression of observed values against independent variable that can be used to predict the missing values—outperformed all other approaches.…”

Section: Resultsmentioning

confidence: 99%

Modeling water demand when households have multiple sources of water

Coulibaly

Jakus

Keith

2014

Water Resources Research

View full text Add to dashboard Cite

A significant portion of the world's population lives in areas where public water delivery systems are unreliable and/or deliver poor quality water. In response, people have developed important alternatives to publicly supplied water. To date, most water demand research has been based on singleequation models for a single source of water, with very few studies that have examined water demand from two sources of water (where all nonpublic system water sources have been aggregated into a single demand). This modeling approach leads to two outcomes. First, the demand models do not capture the full range of alternatives, so the true economic relationship among the alternatives is obscured. Second, and more seriously, economic theory predicts that demand for a good becomes more price-elastic as the number of close substitutes increases. If researchers artificially limit the number of alternatives studied to something less than the true number, the price elasticity estimate may be biased downward. This paper examines water demand in a region with near universal access to piped water, but where system reliability and quality is such that many alternative sources of water exist. In extending the demand analysis to four sources of water, we are able to (i) demonstrate why households choose the water sources they do, (ii) provide a richer description of the demand relationships among sources, and (iii) calculate own-price elasticity estimates that are more elastic than those generally found in the literature.

show abstract

Section: Resultsmentioning

confidence: 99%

Modeling water demand when households have multiple sources of water

Coulibaly

Jakus

Keith

2014

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…This is accomplished through an auxiliary regression in which observed prices for each beverage are regressed on household income, household size, and the region in which the household is located. 2 These variables are used extensively in the literature on imputed prices (Kyureghian, Capps, andNayga 2011, Alviola andCapps 2010). The parameters estimated from the auxiliary regression are then used 1 The Heckman (1979) model speaks only to conditional demand estimates although the irststage probit analysis provides information on the probability of the product being purchased.…”

Section: Methodsmentioning

confidence: 99%

“…to impute prices for the zero-expenditure observations. This technique is wellestablished in the literature and is commonly used to deal with imputing (or forecasting) missing prices and price endogeneity issues (e.g., Capps et al 1994, Alviola and Capps 2010, Kyureghian, Capps, and Nayga 2011, Dharmasena and Capps 2012. We address variability of demand for beverages with different levels of quality via an income variable in the auxiliary regression.…”

Section: Methodsmentioning

confidence: 99%

Unraveling Demand for Dairy-Alternative Beverages in the United States: The Case of Soymilk

Dharmasena

Capps

2014

Agric. resour. econ. rev.

Self Cite

View full text Add to dashboard Cite

Soymilk is one of the fastest growing categories in the U.S dairy alternative functional beverage market. Using household-level purchase data from Nielsen's 2008 Homescan panel and the Tobit econometric procedure, we estimate conditional and unconditional own-price, cross-price, and income elasticities for soymilk, white milk, and flavored milk. Income, age, employment status, education level, race, ethnicity, region, and presence of children in a household are significant drivers of demand for soymilk. White milk and flavored milk are competitors for soymilk, and soymilk is a competitor for white milk. Strategies for pricing and targeted marketing of soymilk are also discussed.

show abstract

“…The use of such imputed prices in the regression model also helps deal with potential endogeneity in the price variable. This approach is not without precedent and is standard procedure in the price imputation literature (Capps et al 1994; Alviola and Capps 2010; Kyureghian, Capps, and Nayga 2011; Dharmasena and Capps 2014). 8 In Table 2, we present summary statistics for observed prices and imputed prices for each of the product categories.…”

Section: Datamentioning

confidence: 99%

Can Dietary Fiber Intake Be Increased through Nutritional Education and through Subsidies on Selected Food Products?

Senia¹,

Dharmasena²,

Capps³

2019

Agric. Resour. Econom. Rev.

Self Cite

View full text Add to dashboard Cite

Consumers in the United States fall short of meeting the recommended guideline for dietary fiber intake. Using a quarterly panel of households from Nielsen for the years 2004 through 2014, we employ a Heckman two-step approach to estimate nine panel regressions concerning per person fiber intakes derived from various food categories to uncover the importance of prices as well as socioeconomic and demographic factors. Prices play a prominent role in the per person intake of dietary fiber derived from the respective food products considered. Households below poverty thresholds had lower intakes of fiber relative to households above poverty thresholds. Ethnicity, race, age of the household head, region, and the presence of children also had significant effects on dietary fiber derived from the respective food categories. A proposed 20 percent subsidy applied to fruits and vegetables would increase per person intake of fiber by 9.4 percent. Therefore, if one were to consider meeting the dietary fiber requirement only through the provision of a subsidy, a large subsidy applied to fruits and vegetables would be required. Therefore, given the complex nature of the various factors affecting the intake of dietary fiber, the feasibility of using subsidies alone to increase the intake of dietary fiber is called into question.

show abstract

A Missing Variable Imputation Methodology with an Empirical Application

Cited by 19 publications

References 5 publications

Modeling water demand when households have multiple sources of water

Modeling water demand when households have multiple sources of water

Unraveling Demand for Dairy-Alternative Beverages in the United States: The Case of Soymilk

Can Dietary Fiber Intake Be Increased through Nutritional Education and through Subsidies on Selected Food Products?

Contact Info

Product

Resources

About