Since the seminal papers of Bates and Granger in 1969, a superfluous amount of information has been published on combining singular forecasts. Materialized evidence has habitually demonstrated that combining the forecasts will produce the best model. Moreover, while it is possible that a best singular model could outperform a composite model, using multiple models provides the advantage of risk diversification. It has also been shown to produce a lower forecasting error. The question to whether to combine has been replaced with what amount of emphasis should be placed on each forecast. Researchers are aspired to derive optimal weights that would produce the lowest forecasting errors. An equal composite of the mean square error, by the covariance, and the best previous model, among others, have been suggested. Other academicians have suggested the use of mechanical derived weights through the use of computer programs. These weights have shown robust results. Once the composite and singular forecasts have been estimated, a systematic approach to evaluate the singular forecasts is needed. Forecasting errors, such as the root mean square error and mean absolute percentage error, are the most common criteria for elimination in both agriculture and other sectors. Although a valid mean of selection, different forecasting errors can produce a different ordinal ranking of the forecasts; thus, producing inconclusive results. These findings have promoted the inspection for other suitable candidates for forecast evaluation. At the forefront of this pursuit is stochastic dominance and stochastic efficiency. Stochastic dominance and stochastic efficiency have traditionally been used as a way to rank wealth or returns from a group of alternatives. They have been principally used in the finance and money sector as a way to evaluate investment strategies. Holt and Brandt in 1985 proposed using stochastic dominance to select between different hedging strategies. Their results suggest that stochastic dominance has the opportunity to feasibly be used in selecting the most accurate forecast. This thesis had three objectives: 1) To determine whether live cattle basis forecasting error could be reduced in comparison to singular models when using composite forecasts 2) To determine whether stochastic dominance and stochastic efficiency could be used to systematically select the most accurate forecasts 3) To determine whether currently reported forecasting error measures might lead to inaccurate conclusions in which forecast was correct. The objectives were evaluated using two primary markets, Utah and Western Kansas, and two secondary markets, Texas and Nebraska. The data for live cattle slaughter steer basis was taken and subsequently computed from the Livestock Marketing Information Center, Chicago Mercantile Exchange, and United States Department of Agriculture from 2004 to 2012. Seven singular were initially used and adapted from the current academic literature. After the models were evaluated using forecasting error, stochastic dominance and stochastic efficiency, seven composite models were created. For each separate composite model, a different weighting scheme was applied. The “optimal” composite weight, in particular, was estimated using GAMS whose objective function was to select the forecast combination that would reduce the variance-covariance between the singular forecasting models. The composite models were likewise systematically evaluated using forecasting error, stochastic dominance and stochastic efficiency. The results indicate that forecasting error can be reduced in all four markets, on the average by using an optimal weighting scheme. Optimal weighting schemes can also outperform the benchmark equal weights. Moreover, a combination of fast reaction time series and market condition, supply and demand, forecasts provide the better model. Stochastic dominance and stochastic efficiency provided confirmatory results and selected the efficient set of the forecasts over a range of risk. It likewise indicated that forecasting error may provide a point estimate rather than a range of error. Suggestions for their application and implementation into extension outlook forecasts and industry application are suggested.

Accurate prediction of Tropical Cyclones (TC) is vital for the protection of life and property in areas that are prone to their destructive forces. While significant improvements have been made in forecasting TC track, intensity remains a challenge. It is hypothesized that accurate TC intensity forecast requires, among other things, an adequate initial description of their inner-core region. Therefore, there must be reliable observations of the inner-core area of the TC and effective data assimilation (DA) methods to ingest these data into the Numerical Weather Prediction (NWP) models. However, these requirements are seldom met at the relatively low resolution of operational global prediction models and the lack of routine observations assimilated in the TC inner-core. This study tests the impacts of assimilating inner-core Airborne Doppler Radar (ADR) winds on the initial structure and subsequent intensity forecast of Hurricane Ike (2008). The 4-dimensional variational (4DVar) and the 3-dimensional variational (3DVar) methods are used to perform DA while the Weather Research and Forecasting (WRF) model is used to perform forecasts. It is found that assimilating data helps to initialize a more realistic inner-core structure using both DA methods. Additionally, the resulting short-term and long-term intensity forecasts are more accurate when data is assimilated versus cases when there is no DA. Additionally, it is found that in some cases the impact of DA lasts up to 12 hours longer with 4DVar versus 3DVar. It is shown that this is because the flow-dependent 4DVar method produces more dynamically and balanced analysis increments compared to the static and isotropic increments of 3DVar. However, the impact of using both methods is minimal in the long-range. The analyses show that at longer forecast range the dynamics of hurricane Ike was influenced more by outer environment features than the inner-core winds.

This dissertation investigates the relation between firm disclosure, analyst forecast bias, and the cost of equity capital (COEC). Since analyst forecast bias is associated with both implied COEC estimates and disclosure, it is important to control for or remove it from COEC estimates when estimating the relation between disclosure and ex ante expected returns. I begin my analysis by predicting and removing systematic ex ante bias from analyst forecasts to produce de-biased analyst forecasts that better proxy for the market’s ex ante earnings expectations. I use these de-biased analyst forecasts to produce estimates of ex ante expected returns, both at the portfolio- and the firm-level. In addition, I develop a novel estimate of ex ante expected returns by applying Vuolteenaho’s (2002) return decomposition framework to ex post realized returns and accounting data. Finally, using several techniques to control for analyst forecast bias and self-selection bias, I find theoretically consistent evidence of a negative association between regular disclosure and ex ante expected returns. I predict and show that inferences can change when analyst forecast bias is controlled for.

This dissertation investigates the relation between firm disclosure, analyst forecast bias, and the cost of equity capital (COEC). Since analyst forecast bias is associated with both implied COEC estimates and disclosure, it is important to control for or remove it from COEC estimates when estimating the relation between disclosure and ex ante expected returns. I begin my analysis by predicting and removing systematic ex ante bias from analyst forecasts to produce de-biased analyst forecasts that better proxy for the market’s ex ante earnings expectations. I use these de-biased analyst forecasts to produce estimates of ex ante expected returns, both at the portfolio- and the firm-level. In addition, I develop a novel estimate of ex ante expected returns by applying Vuolteenaho’s (2002) return decomposition framework to ex post realized returns and accounting data. Finally, using several techniques to control for analyst forecast bias and self-selection bias, I find theoretically consistent evidence of a negative association between regular disclosure and ex ante expected returns. I predict and show that inferences can change when analyst forecast bias is controlled for.

No description available.

Prior research has established that stocks with high dispersion of earnings forecasts or short interest are associated with low subsequent returns. Assuming dispersion of forecasts is a proxy for divergence of opinions and short interest is a proxy for short selling constraints, these results have been traditionally attributed to correction for overpricing created by binding short selling constraints. This argument is provided by Miller (1977), and states that prices reflect an optimistic view when investors with pessimistic views can not trade due to short selling constraints, and that the more opinions diverge, the more stocks become overpriced. I test whether dispersion of forecasts exacerbates overpricing, but find evidence contrary to Miller’s theory. When dispersion of forecasts increases, prices decrease. I offer an explanation based on analysts’ reluctance to quickly revise their forecasts downward. I show that some analysts’ sluggish response to bad news results in dispersion of forecasts. The inertia in downward forecast revisions also leads to market underreaction to bad news. Therefore, the negative relationship between dispersion and subsequent returns may be attributable to analysts’ sluggish response to bad news. I also examine the return predictability of firms with high short interest and low institutional ownership. Short interest seems to predict not only future stock returns but also future earnings news, especially for firms with lower institutional ownership. Therefore, the return predictability of short interest seems to be associated with value relevant information short sellers seem to have gathered.

This study investigates whether auditors respond to earnings management pressure created by analyst forecasts. Analyst forecasts create an important earnings target for management, and professional standards direct auditors to consider how this pressure could affect their clients. Using annual analyst forecasts available during the planning phase of the audit, I examine whether this form of earnings management pressure affects clients’ financial statement misstatements. Next, I investigate whether auditors respond to earnings forecast pressure through audit fees and reporting delay. I find that higher levels of analyst forecast pressure increase the likelihood of client restatement. I also find that auditors charge higher audit fees and delay the issuance of the audit report in response to pressure from analyst expectations. Finally, I find that when audit clients are subject to high analyst forecast pressure, a high audit fee response by auditors mitigates the likelihood of client misstatements.

Forecasting firms' earnings has long been an interest of market participants and academics. Traditional forecasting studies in a multivariate time series setting do not take into account that the timing of market data release for a specific time period of observation is often spread over several days or weeks. This thesis focuses on the separation of announcement timing or data release and the use of econometric real-time methods, which we refer to as an updated vector autoregression (VAR) forecast, to predict data that have yet to be released. In comparison to standard time series forecasting, we show that the updated forecasts will be more accurate the higher the correlation coefficients among the standard VAR innovations are. Forecasting with the sequential release of information has not been studied in the VAR framework, and our approach to U.S. nonfarm payroll employment and the six Canadian banks shows its value. By using the updated VAR forecast, we conclude that there are relative efficiency gains in the one-step-ahead forecast compared to the ordinary VAR forecast, and compared to professional consensus forecasts. Thought experiments emphasize that the release ordering is crucial in determining forecast accuracy. / Thesis (Ph.D, Economics) -- Queen's University, 2010-04-30 12:34:42.629

Comparability is a desirable qualitative characteristic of financial information and critical for financial statement users' ability to identify and understand similarities and differences in financial results among reporting entities. Yet, little research explicitly considers either the determinants or benefits of comparability because of difficulty in identifying and measuring the theoretical construct of comparability. Further, the widespread global adoption of IFRS, a relatively homogenous set of accounting standards, is expected to increase comparability among companies that operate in different national jurisdictions. However, prior studies that examine the average impact of mandatory IFRS adoption on comparability find mixed results. I hypothesized that the impact of mandatory IFRS adoption on comparability varies with managers' reporting incentives and differences between countries' domestic standards and IFRS. Using listed firms from 34 countries, I documented that comparability under non-IFRS domestic standards is higher in countries that provide strong reporting incentives (i.e. countries with strict enforcement regimes or high earnings transparency). Additionally, I found an increase in comparability following IFRS adoption (relative to a control sample of non-adopters) in countries that provide strong reporting incentives or with large domestic GAAP-IFRS differences. In contrast, I found evidence of a decrease following IFRS adoption (relative to a control sample of non-adopters) in countries with weak reporting incentives or with small domestic GAAP-IFRS differences. Finally, I showed that changes in comparability surrounding adoption are positively associated with changes in the quality of firms' information environments.

An empirical method is developed for constructing likelihood functions required in a Bayesian probabilistic flash flood forecasting model using data on objective quantitative precipitation forecasts and their verification. Likelihoods based on categorical and probabilistic forecast information for several forecast periods, seasons, and locations are shown and compared. Data record length, forecast information type and magnitude, grid area, and discretized interval size are shown to affect probabilistic differentiation of amounts of potential rainfall. Use of these likelihoods in Bayes' Theorem to update prior probability distributions of potential rainfall, based on preliminary data, to posterior probability distributions, reflecting the latest forecast information, demonstrates that an abbreviated version of the flash flood forecasting methodology is currently practicable. For this application, likelihoods based on the categorical forecast are indicated. Apart from flash flood forecasting, it is shown that likelihoods can provide detailed insight into the value of information contained in particular forecast products.