Analytic Derivatives for Estimation of Linear Dynamic Models
November 1988
Working Paper Number:
CES-88-05
Abstract
This paper develops two algorithms. Algorithm I computes the exact, Gaussian, log-likelihood function, its exact, gradient vector, and an asymptotic approximation of its Hessian matrix, for discrete-time, linear, dynamic models in state-space form. Algorithm 2, derived from algorithm I, computes the exact, sample, information matrix of this likelihood function. The computed quantities are analytic (not numerical approximations) and should, therefore, be useful for reliably, quickly, and accurately: (i) checking local identifiability of parameters by checking the rank of the information matrix; (ii) using the gradient vector and Hessian matrix to compute maximum likelihood estimates of parameters with Newton methods; and, (iii) computing asymptotic covariances (Cramer-Rao bounds) of the parameter estimates with the Hessian or the information matrix. The principal contribution of the paper is algorithm 2, which extends to multivariate models the univariate results of Porat and Friedlander (1986). By relying on the Kalman filter instead of the Levinson-Durbin filter used by Porat and Friedlander, algorithms 1 and 2 can automatically handle any pattern of missing or linearly aggregated data. Although algorithm 1 is well known, it is treated in detail in order to make the paper self contained.
Document Tags and Keywords
Keywords
Keywords are automatically generated using KeyBERT, a powerful and innovative
keyword extraction tool that utilizes BERT embeddings to ensure high-quality and contextually relevant
keywords.
By analyzing the content of working papers, KeyBERT identifies terms and phrases that capture the essence of the
text, highlighting the most significant topics and trends. This approach not only enhances searchability but
provides connections that go beyond potentially domain-specific author-defined keywords.
:
Tags
Tags are automatically generated using a pretrained language model from spaCy, which excels at
several tasks, including entity tagging.
The model is able to label words and phrases by part-of-speech,
including "organizations." By filtering for frequent words and phrases labeled as "organizations", papers are
identified to contain references to specific institutions, datasets, and other organizations.
:
Similar Working Papers
Similarity between working papers are determined by an unsupervised neural
network model
know as Doc2Vec.
Doc2Vec is a model that represents entire documents as fixed-length vectors, allowing for the
capture of semantic meaning in a way that relates to the context of words within the document. The model learns to
associate a unique vector with each document while simultaneously learning word vectors, enabling tasks such as
document classification, clustering, and similarity detection by preserving the order and structure of words. The
document vectors are compared using cosine similarity/distance to determine the most similar working papers.
Papers identified with 🔥 are in the top 20% of similarity.
The 10 most similar working papers to the working paper 'Analytic Derivatives for Estimation of Linear Dynamic Models' are listed below in order of similarity.
-
Working PaperThe Modern Wholesaler: Global Sourcing, Domestic Distribution, and Scale Economies
December 2018
Working Paper Number:
CES-18-49
Nearly half of all transactions in the $6 trillion market for manufactured goods in the United States were intermediated by wholesalers in 2012, up from 32 percent in 1992. Seventy percent of this increase is due to the growth of 'superstar' firms - the largest one percent of wholesalers. Structural estimates based on detailed administrative data show that the rise of the largest wholesalers was driven by an intuitive linkage between their sourcing of goods from abroad and an expansion of their domestic distribution network to reach more buyers. Both elements require scale economies and lead to increased wholesaler market shares and markups. Counterfactual analysis shows that despite increases in wholesaler market power, intermediated international trade has two benefits for buyers: directly through buyers' valuation of globally sourced products, and indirectly through the passed-through benefits of wholesaler economies of scale and increased quality.View Full Paper PDF
-
Working PaperThe Choice of Input-Out Table Embedded in Regional Econometric Input-Out Models
January 1994
Working Paper Number:
CES-94-01
In this paper we investigate the role of input-output data source in the regional econometric input-output models. While there has been a great deal of experimentation focused on the accuracy of alternative methods for estimating regional input-output coefficients, little attention has been directed to the role of accuracy when the input-output system is nested within a broader accounting framework. The issues of accuracy were considered in two contexts, forecasting and impact analysis focusing on a model developed for the Chicago Region. We experimented with three input-output data sources: observed regional data, national input-output, and randomly generated input-output coefficients. The effects of different sources of input-output data on regional econometric input-output model revealed that there are significant differences in results obtained in impact analyses. However, the adjustment processes inherent in the econometric input-output system seem to mute the initial differences in input- output data when the model is used for forecasting. Since applications of these types of models involve both impact and forecasting exercises, there would still seem to be a strong motivation for basing the system on the most accurate set of input-output accounts.View Full Paper PDF
-
Working PaperScale Economies and Consolidation in Hog Slaughter
March 2000
Working Paper Number:
CES-00-03
We use establishment based panel data to estimate a cost function which identifies the role of scale economies in hog slaughter consolidation. We find modest by extensive technological scale economies in the 1990s, and they became more important over time. But wages rose sharply with plant size through the 1970s and those wage premiums generated a pecuniary scale diseconomy that largely offset the effects of technological scale economies. The size-wage relation disappeared in the 1980; with growing technological scale economies and disappearing pecuniary diseconomies, large plants realized growing cost advantages over smaller plants, and production shifted to larger plants.View Full Paper PDF
-
Working PaperEconomic Factors Underlying the Unbundling of Advertising Agency Services
August 2009
Working Paper Number:
CES-09-15
This paper addresses a longstanding puzzle involving the unbundling of services that has occurred over more than two decades in the U.S. advertising agency industry: How can the shift from the bundling to the unbundling of services be explained and what accounts for the slow pace of change? Using a cost-based theoretical framework of bundling due to Evans and Salinger (2005, 2008), we develop a simple model of an advertising agency's decision to unbundle its services as a tradeoff between the fixed cost to the advertiser of establishing and maintaining a relationship with an advertising agency and pecuniary economies of scale available in providing media services. The results from an econometric analysis of cross-sectional and pooled data collected by the U.S. Census Bureau for quinquenial censuses conducted between 1982 and 2002 support the key predictions of the model. We find that advertising agency establishments are more likely to unbundle if they are large and diversified in their service offerings and are less likely to do so with increasing age and greater geographical scope. We also find a strong trend toward unbundling over time, a result that is partially explained by increases in media prices over time.View Full Paper PDF
-
Working PaperEstimating the Impact of Low-Income Universal Service Programs
June 2013
Working Paper Number:
CES-13-33
This policy study uses U.S. Census microdata to evaluate how subsidies for universal telephone service vary in their impact across low-income racial groups, gender, age, and home ownership. Our demand specification includes both the subsidized monthly price (Lifeline program) and the subsidized initial connection price (Linkup program) for local telephone service. Our quasimaximum likelihood estimation controls for location differences and instruments for price endogeneity. The microdata allow us to estimate the effects of demographics on both elasticities of telephone penetration and the level of telephone penetration. Based on our preferred estimates, the subsidy programs increased aggregate penetration by 6.1% for low-income households. Our results suggest that Linkup is more cost-effective than Lifeline and that auto-enroll policies are important, which calls into question a recent FCC (2012) decision to reduce Linkup subsidies in favor of Lifeline. Our study can inform the evaluation of similar universal service policies for Internet access.View Full Paper PDF
-
Working PaperOutsourcing Business Service and the Scope of Local Markets
August 2000
Working Paper Number:
CES-00-14
This paper examines outsourcing to test whether productivity-enhancing specialization is facilitated in bigger cities. First, the paper provides a theoretical model which shows that greater local demand for a given input promotes the entry of suppliers into a city; the increased number of suppliers then results in lower outsourcing prices and a higher use of outsourcing by final producers, therefore reducing the final producers' production costs. I then test the predictions of the model by examining manufacturing plants' practices of outsourcing business services, by using plant-level data from the 1992 Annual Survey of Manufactures. The empirical results show that an exogenous increase in local demand promotes the entry of service suppliers and increases a firm's probability of outsourcing for white-collar services. In particular, I found that doubling the intensity of the use of a service in a U.S. county, which can be attributed to the industrial composition of the county, results in a 7% to 25% increase in the probability of outsourcing.View Full Paper PDF
-
Working PaperWhat Drives Differences in Management?
January 2017
Working Paper Number:
CES-17-32
Partnering with the Census we implement a new survey of 'structured' management practices in 32,000 US manufacturing plants. We find an enormous dispersion of management practices across plants, with 40% of this variation across plants within the same firm. This management variation accounts for about a fifth of the spread of productivity, a similar fraction as that accounted for by R&D and twice as much as explained by IT. We find evidence for four 'drivers' of management: competition, business environment, learning spillovers and human capital. Collectively, these drivers account for about a third of the dispersion of structured management practices.View Full Paper PDF
-
Working PaperReallocation and Technology: Evidence From The U.S. Steel Industry
March 2013
Working Paper Number:
CES-13-06
We measure the impact of a drastic new technology for producing steel -- the minimill -- on the aggregate productivity of U.S. steel producers, using unique plant-level data between 1963 and 2002. We find that the sharp increase in the industry's productivity is linked to this new technology, and operates through two distinct mechanisms. First, minimills displaced the older technology, called vertically integrated production, and this reallocation of output was responsible for a third of the increase in the industry's productivity. Second, increased competition, due to the expansion of minimills, drove a substantial reallocation process within the group of vertically integrated producers, driving a resurgence in their productivity, and consequently of the industry's productivity as a whole.View Full Paper PDF
-
Working PaperThe Slow Growth of New Plants: Learning about Demand?
March 2012
Working Paper Number:
CES-12-06
It is well known that new businesses are typically much smaller than their established industry competitors, and that this size gap closes slowly. We show that even in commodity-like product markets, these patterns do not reflect productivity gaps, but rather differences in demand-side fundamentals. We document and explore patterns in plants' idiosyncratic demand levels by estimating a dynamic model of plant expansion in the presence of a demand accumulation process (e.g., building a customer base). We find active accumulation driven by plants' past production decisions quantitatively dominates passive demand accumulation, and that within-firm spillovers affect demand levels but not growth.View Full Paper PDF
-
Working PaperThe Agglomeration of Headquarters
February 2004
Working Paper Number:
CES-04-02
This paper uses a micro data set on auxiliary establishments from 1977 to 1997 in order to investigate the determinants of headquarter agglomerations and the underlying economic base of many larger metro areas. The significance of headquarters in large urban settings is their ability to facilitate the spatial separation of their white collar activities from remote production plants. The results show that separation benefits headquarters in two main ways: the availability of di?erentiated local service input suppliers and the scale of other headquarter activity nearby. A wide diversity of local service options allows the headquarters to better match their various needs with specific experts producing service inputs from whom they learn, which improves their productivity. Headquarters also benefit from other headquarter neighbors, although such marginal scale benefits seem to diminish as local scale rises.View Full Paper PDF
