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 PaperFirms Started As Franchises Have Lower Survival Rates Than Independent Small Business Startups
May 1994
Working Paper Number:
CES-94-03
Aspiring entrepreneurs choosing to become franchisees certainly expect to improve their chances of survival during the turbulent early years of business startup and operation. Alignment with a franchiser parent company offers the franchisee managerial assistance, access to financial capital, and access to markets via the right to utilize the parent company trademark. This study examines survival patterns among franchise and nonfranchise small firms started between 1984 and 1987: survival through late 1991 is tracked for all firms. Although the franchise operations are larger scale, better capitalized young firms, the independent business startups are found to be more profitable and their survival prospects are better than those of franchises.View Full Paper PDF
-
Working PaperHuman Capital Spillovers in Manufacturing: Evidence from Plant-Level Production Functions
November 2002
Working Paper Number:
CES-02-27
I assess the magnitude of human capital spillovers in US cities by estimating plant level production functions. I use a unique firm-worker matched dataset, obtained by combining the Census of Manufacturers with the Census of Population. After controlling for a plant's own human capital, plant fixed effects, industry-specific and state-specific transitory shocks, I find that the output of plants located in cities that experience large increases in the share of college graduates rises more than the output of similar plants located in cities that experience small increases in the share of college graduates. Several specification tests indicate that the estimated effect is not completely spurious. First, within a city, the spillover between plants that are geographically and economically close is positive, while spillovers between plants that are geographically close but economically distant is zero. Second, most of the estimated spillover comes from hi-tech plants. For non hi-tech productions, the spillover is virtually zero. When I stratify the sample by the percentage of employees who are college educated, I find that the spillover is larger the larger the percentage of college educated workers in the plant. Third, density of physical capital in a city outside a plant has no effect on a plant's productivity. Consistent with a model that includes both standard and general equilibrium forces and spillovers, the estimated productivity differences between cities with high and low levels of human capital match remarkably well differences in labor costs that are typically observed between cities with high and low levels of human capital. This is important because, in equilibrium, any productivity gain generated by human capital spillover should be offset by increased costs.View Full Paper PDF
-
Working PaperWhy Are Plant Deaths Countercyclical: Reallocation Timing or Fragility?
November 2006
Working Paper Number:
CES-06-24
Because plant deaths destroy specific capital with large local economic impacts and potentially important macroeconmic effects, understanding the causes of deaths and, in particular, why they are concentrated in cyclical downturns, is important. The reallocationtiming hypothesis posits that plants suffering adverse permanent demand/productivity shocks delay shutdowns until cyclical downturns when plant capacity is less valuable, while the fragility hypothesis posits that shutdowns occur in downturns because the option value of maintaining the plant through low profitability periods is too small. I show that the effect that a plant's specific capital has on the timing of plant deaths differs across these two hypotheses and then use this insight to test the hypotheses' relative importance. I find that fragility is the dominant cause of the countercyclical behavior of plant deaths. This suggests that the endogenous destruction of capital is likely an important amplification and propagation mechanism for cyclical shocks and that stabilization policies have the benefit of reduced capital destruction.View Full Paper PDF
-
Working PaperIT Investment and Firm Performance in U.S. Retail Trade
June 2002
Working Paper Number:
CES-02-14
We examine the relationships between investments in information technology (IT) and two measures of retail firm performance -- productivity and establishment growth -- over the 1992 to 1997 period. We use untapped firm and establishment micro data from the Censuses of Retail Trade and the Assets and Expenditures Survey. We show that large firms account for most retail IT investment, employment and establishment growth. We find evidence of a significant relationship between IT investment intensity and productivity growth. We found no such evidence of a link between IT growth in the number of establishments operated by retail firms.View Full Paper PDF
-
Working PaperGeographic Concentration as a Dynamic Process
March 1998
Working Paper Number:
CES-98-03
This degree of geographic concentration of individual manufacturing industries in the U.S. has declined only slightly in the last twenty years. At the same time, new plant births, plant expansions, contractions and closures have shifted large quantities of employment across plants, firms and locations. This paper uses data from the Census Bureau's Longitudinal Research Database to examine how relatively stable levels of geographic concentration emerge from this dynamic process. While industries agglomeration levels tend to remain fairly constant we find that there is a greater variation in the locations of these agglomerations. We then decompose aggregate concentration changes into portions attributable to plant births, expansions, contractions, and closures, and find that the location choices of new firms and differences in growth rates have played the most significant role in reducing levels of geographic concentration, while plant closures have tended to reinforce agglomeration. Finally, we look at coagglomeration patterns to test three of Marshall's theories of industry agglomeration: (1) agglomeration saves transport costs by proximity to input suppliers or final consumers, (2) agglomeration allows for labor market pooling, and (3) agglomeration facilitates intellectual spillovers. While there is some truth behind all three theories, we find that industrial location is far more driven by labor mix than by any of the other explanatory variables.View Full Paper PDF
-
Working PaperDoes Firms' Financial Status Affect Plant-Level Investment and Exit Decisions?
January 1999
Working Paper Number:
CES-99-03
This paper investigates the influence of a firm's financial status on the within-firm allocation of funds, reflected in its plant-level investment and exit decisions. In the empirical analysis, financial status is measured by both standard measures and an indicator variable recently suggested by Kaplan and Zingales. Based on these firm-level financial variables and on planet-level investment and production data from the U.S. Census Bureau's Longitudinal Research Database(LRD), econometric models of plant operating regimes are estimated which summarize investment and exit decisions. The empirical evidence supports the view that firm-level financial status affects investments and market exit decisions observed at the plant level.View Full Paper PDF
-
Working PaperTechnifying Ventures
July 2025
Working Paper Number:
CES-25-49
How do advanced technology adoption and venture capital (VC) funding impact employment and growth? An analysis of data from the US Census Bureau suggests that while both advanced technology use and VC funding matter on their own for firm outcomes, their joint presence is most strongly correlated with higher employment levels. VC presence is linked with a high increase in employment, though primarily among a limited subset of firms. In contrast, technology adoption is associated with a smaller rise in employment, yet it influences a considerably larger number of firms. A model of startups is created, focusing on decisions to use advanced technology and seek VC funding. The model is compared with firm-level data on employment, advanced technology use, and VC investment. Several thought experiments are conducted using the model. Some experiments assess the importance of advanced technology and VC in the economy. Others examine the reallocation effects across firms with different technology choices and funding sources in response to shifts in taxes and subsidies.View Full Paper PDF
-
Working PaperHow Does Labor Market Size Affect Firm Capital Structure? Evidence from Large Plant Openings
November 2015
Working Paper Number:
CES-15-38
I examine how the labor market in which firms operate affects their capital structure decisions. Using the US Census Bureau data, I exploit a large plant opening as an abrupt increase in the size of a local labor market. I find that a new plant opening leads to a 2.6% to 3.9% increase in the debt-to-capital ratio of existing firms in the 'winner' county relative to the 'runner-up' choice. This result is consistent with larger labor markets making a job loss less costly, which in turn reduces indirect costs of financial distress. Moreover, this spillover effect is larger for firms 1) that have a larger fraction of employees in the affected county, 2) that employ the same type of workers as the new plant, and 3) that have larger unexploited benefits of debt.View Full Paper PDF
-
Working PaperThe Longitudinal Business Database
July 2002
Working Paper Number:
CES-02-17
As the largest federal statistical agency and primary collector of data on businesses, households and individuals, the Census Bureau each year conducts numerous surveys intended to provide statistics on a wide range of topics about the population and economy of the United States. The Census Bureau's decennial population and quinquennial economic censuses are unique, providing information on all U.S. households and business establishments, respectively.View Full Paper PDF
-
Working PaperThe United States Labor Market: Status Quo or A New Normal?
September 2012
Working Paper Number:
CES-12-28
The recession of 2007-09 witnessed high rates of unemployment that have been slow to recede. This has led many to conclude that structural changes have occurred in the labor market and that the economy will not return to the low rates of unemployment that prevailed in the recent past. Is this true? The question is important because central banks may be able to reduce unemployment that is cyclic in nature, but not that which is structural. An analysis of labor market data suggests that there are no structural changes that can explain movements in unemployment rates over recent years. Neither industrial nor demographic shifts nor a mismatch of skills with job vacancies is behind the increased rates of unemployment. Although mismatch increased during the recession, it retreated at the same rate. The patterns observed are consistent with unemployment being caused by cyclic phenomena that are more pronounced during the current recession than in prior recessions.View Full Paper PDF
