FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2188458215> ?p ?o ?g. }

• W2188458215 abstract "Marketers have always struggled to accurately determine the lagged effects of marketing. Correctly estimating marketing contributions is another important challenge faced by marketing professionals. In an earlier paper, we showed that a generalized additive model (using PROC GAM) accurately estimates the lagged effects of advertis- ing efforts in simulated models with one independent variable. In this paper we take this research a step further, and perform simulations with additional independent variables to examine whether GAM is still able to accurately estimate lagged effects. We also look at the calculation of marketing contributions to determine the accuracy of those estimates. This will allow the end user to determine if the promising results obtained earlier will carry over in real world applications. While much of the literature has been devoted to various ways of handling the lags, there have been very few pa- pers that have looked at how an incorrect model specification can affect the estimation of advertising carryover effects. In an earlier paper we pointed out that incorrect model specification can lead to errors in the estimation of advertising lags. We proposed that there are two ways in which incorrect specifications creep into the models in practice. First, when it comes to estimations of dynamic lags, assumptions are made about the relationships be- tween the parameter estimates of the lagged independent variables. For instance, in the geometric lag model, the parameters of the lagged variables are assumed to be geometrically declining over time. Similarly, in the Pascal model, the coefficients of the lagged terms are assumed to follow a negative binomial distribution. The very popu- lar Polynomial distributed lagged model (Proc PDLREG) also assumes that the lag coefficients lie on a polynomial curve. Since in reality, the coefficients of the lagged terms may not be related in any of the above ways, a model fitted under any of those assumptions may be mis-specified. The second factor that complicates matters is that, often, the relationship between the dependent and independent variable may be nonlinear. Therefore, in order to simplify the estimation process, assumptions are often made about the relationships between the dependent and independent variables. For instance, when we know the relationship between the dependent and independent var- iable is non-linear we may use a specific functional form (such as logarithmic or square root or reciprocal) to mod- el that relationship. (For more details about the types of functional forms that are usually used, please refer to the NESUG 2010 paper by the same author). However, rarely in the real world does the relationship between sales and advertising follow a specific mathemati- cal functional form. Moreover, predictor variables usually do not show much variation in the sample. Therefore we may only observe values of advertising within a small range. Sometimes with only small variations in the sample, several models can be a good fit for the data. Since real world data is unlikely to adhere to a specific functional form, using any particular function to model the data may lead to misspecification problems. In previous papers we have used simulation methods to look at the role of model specification in the estimation of advertising lags. We found that when the true model is different from the model that is specified, incorrect adver- tising lags will often show up as significant in the model. We found that two kinds of errors can occur when an in-" @default.
• W2188458215 created "2016-06-24" @default.
• W2188458215 creator A5049588964 @default.
• W2188458215 date "2012-01-01" @default.
• W2188458215 modified "2023-09-27" @default.
• W2188458215 title "GAM in Marketing Mix Modeling: Revisited" @default.
• W2188458215 cites W1998860838 @default.
• W2188458215 cites W2072459128 @default.
• W2188458215 cites W2157448961 @default.
• W2188458215 cites W2162187820 @default.
• W2188458215 cites W2183095417 @default.
• W2188458215 cites W2315189750 @default.
• W2188458215 cites W2329480876 @default.
• W2188458215 cites W2797583072 @default.
• W2188458215 cites W3014310718 @default.
• W2188458215 hasPublicationYear "2012" @default.
• W2188458215 type Work @default.
• W2188458215 sameAs 2188458215 @default.
• W2188458215 citedByCount "0" @default.
• W2188458215 crossrefType "journal-article" @default.
• W2188458215 hasAuthorship W2188458215A5049588964 @default.
• W2188458215 hasConcept C105795698 @default.
• W2188458215 hasConcept C117222624 @default.
• W2188458215 hasConcept C134306372 @default.
• W2188458215 hasConcept C144133560 @default.
• W2188458215 hasConcept C149782125 @default.
• W2188458215 hasConcept C162324750 @default.
• W2188458215 hasConcept C162853370 @default.
• W2188458215 hasConcept C182365436 @default.
• W2188458215 hasConcept C187736073 @default.
• W2188458215 hasConcept C27574286 @default.
• W2188458215 hasConcept C31258907 @default.
• W2188458215 hasConcept C33923547 @default.
• W2188458215 hasConcept C41008148 @default.
• W2188458215 hasConcept C75778745 @default.
• W2188458215 hasConcept C96250715 @default.
• W2188458215 hasConcept C98716237 @default.
• W2188458215 hasConceptScore W2188458215C105795698 @default.
• W2188458215 hasConceptScore W2188458215C117222624 @default.
• W2188458215 hasConceptScore W2188458215C134306372 @default.
• W2188458215 hasConceptScore W2188458215C144133560 @default.
• W2188458215 hasConceptScore W2188458215C149782125 @default.
• W2188458215 hasConceptScore W2188458215C162324750 @default.
• W2188458215 hasConceptScore W2188458215C162853370 @default.
• W2188458215 hasConceptScore W2188458215C182365436 @default.
• W2188458215 hasConceptScore W2188458215C187736073 @default.
• W2188458215 hasConceptScore W2188458215C27574286 @default.
• W2188458215 hasConceptScore W2188458215C31258907 @default.
• W2188458215 hasConceptScore W2188458215C33923547 @default.
• W2188458215 hasConceptScore W2188458215C41008148 @default.
• W2188458215 hasConceptScore W2188458215C75778745 @default.
• W2188458215 hasConceptScore W2188458215C96250715 @default.
• W2188458215 hasConceptScore W2188458215C98716237 @default.
• W2188458215 hasLocation W21884582151 @default.
• W2188458215 hasOpenAccess W2188458215 @default.
• W2188458215 hasPrimaryLocation W21884582151 @default.
• W2188458215 hasRelatedWork W121902686 @default.
• W2188458215 hasRelatedWork W1584029445 @default.
• W2188458215 hasRelatedWork W1608254561 @default.
• W2188458215 hasRelatedWork W1974858978 @default.
• W2188458215 hasRelatedWork W2025579466 @default.
• W2188458215 hasRelatedWork W2049850646 @default.
• W2188458215 hasRelatedWork W2101943658 @default.
• W2188458215 hasRelatedWork W2188789984 @default.
• W2188458215 hasRelatedWork W2329990193 @default.
• W2188458215 hasRelatedWork W2331026048 @default.
• W2188458215 hasRelatedWork W245810865 @default.
• W2188458215 hasRelatedWork W2617487884 @default.
• W2188458215 hasRelatedWork W3022606520 @default.
• W2188458215 hasRelatedWork W3121209109 @default.
• W2188458215 hasRelatedWork W3122688920 @default.
• W2188458215 hasRelatedWork W3122839390 @default.
• W2188458215 hasRelatedWork W3209465335 @default.
• W2188458215 hasRelatedWork W585551332 @default.
• W2188458215 hasRelatedWork W597440628 @default.
• W2188458215 hasRelatedWork W1968159702 @default.
• W2188458215 isParatext "false" @default.
• W2188458215 isRetracted "false" @default.
• W2188458215 magId "2188458215" @default.
• W2188458215 workType "article" @default.