Rational Forecasts Research Articles

Справочная информационная система FLR-Library для адаптивного лесовосстановления: кластерный анализ дескрипторов

The lack of uniform terminological and technological approaches to the procedure of adaptive reforestation suggests the justification of the principles of creating a reference information system (FLR-Library), taking into account the specifics of implementing a joint decomposed FLR algorithm to reduce its execution time and rational forecasting of future forest landscape restoration projects. Collected and analyzed more than 120 definitions for more than 30 descriptors, such as "forest", "forest landscapes", "adaptive reforestation", "concepts of adaptive reforestation", "reforestation", "afforestation", "stand", "forest plantation", "forest complex", "forest landscape", "type of forest" and others. Both distinctive features and common points in the definitions of different countries and authors of the same descriptor are found. Technically, an approach to clustering of the most common descriptors is implemented: "forest", "forest landscapes", "reforestation", "afforestation". To do this, for each descriptor, based on the hierarchical clustering module, the combination of a cluster of recalibrated distances was determined and a cluster diagram was constructed based on «variables» (criteria for the descriptors effectiveness) and «observations» (by countries containing a source of information with a descriptor). Common to all sources will be that adaptive reforestation implies adaptation to climate change, but then there are two options for development: 1) planting (sowing) breeds resistant to climate change; 2) import and adaptation of "southern" breeds to northern latitudes. In the future, based on the analysis of descriptors and the principle of decomposition, it is planned to develop an algorithm for the functioning of the FLR system.

Modeling Multi-horizon Electricity Demand Forecasts in Australia: A Term Structure Approach

The Australian Electricity Market Operator generates one-day ahead electricity demand forecasts for the National Electricity Market in Australia and updates these forecasts over time until the time of dispatch. Despite the fact that these forecasts play a crucial role in the decision-making process of market participants, little attention has been paid to their evaluation and interpretation. Using half-hourly data from 2011 to 2015 for New South Wales and Queensland, it is shown that the official half-hourly demand forecasts do not satisfy the econometric properties required of rational forecasts. Instead there is a relationship between forecasts and forecast horizon similar to a term structure model of interest rates. To study the term structure of demand forecasts, a factor analysis that uses a small set of latent factors to explain the common variation among multiple observables is implemented. A three-factor model is identified with the factors admitting interpretation as the level, slope and curvature of the term structure of forecasts. The validity of the model is reinforced by assessing the economic value of demand forecasts. It is demonstrated that simple adjustments to long-horizon electricity demand forecasts based on the three estimated factors can enhance the informational content of the official forecasts.

Are internally consistent forecasts rational?

AbstractEconomic forecasts, such as the Survey of Professional Forecasters (SPF), are revised multiple times before realization of the target. This paper studies the sources of forecast revisions. By decomposing the fixed‐event forecast error into a rational component due to unanticipated future shocks and an irrational component due to measurement error in acquired information or forecasters' reactions, we derive the conditions under which fixed‐event forecasts that contain an irrational forecast error can still possess the second moment properties of rational forecasts. We show that internal consistency of fixed‐event forecasts depends on the magnitude of the rational and irrational components in the revisions. As such, fixed‐event forecasts subject to irrational error may still be internally consistent, although they are not rational, as evidence in many empirical studies. We illustrate our methodology with the SPF inflation forecasts data. Our results show evidence of a sizeable and heterogeneous irrational forecast error component across forecast horizons and a high irrational‐to‐news ratio for most forecasters. This finding also provides insight into why forecast revision effort is not always fully compensated by revision reward.

Assessing the Informational Content of Official Australian Bureau of Meteorology Forecasts of Wind Speed*

Understanding forecast revisions is critical for weather forecast users to determine the optimal timing for their planning decisions. A set of multi‐horizon forecasts for wind speed produced by the Australian Bureau of Meteorology for 12 weather stations in eastern Australia are examined. The forecasts are examined in terms of the econometric definition of rationality and, as a robustness check, the economic value of the forecasts is also assessed using a cost–loss model. It is demonstrated that while the forecasts exhibit some of the characteristics of rational forecasts, when official testing is introduced forecast rationality is rejected at all the weather stations considered. Furthermore, the behaviour of the forecasts is shown to be very erratic over the course of the day and over forecast horizons. Although there is some evidence that the official forecasts can provide positive economic value, this metric also indicates that there is substantial room for improvement.

Gasoline price predictability: insights from consumer vehicle‐buying assessments

AbstractThis study employs the random walk forecasts of US retail gasoline prices for 2001–2020 in order to investigate the potential usefulness of consumer vehicle‐buying assessments for predicting gasoline prices. We show that the random walk forecasts are free of systematic bias and imply symmetric loss, but they do not fully contain the information on consumer vehicle‐buying assessments. For instance, the random walk forecast errors fail to be orthogonal to the difference of opinion of consumers about whether the next 12 months or so will be a good time or a bad time to buy a new vehicle, and to the difference of opinion of consumers about vehicle prices (but not interest rates). In addition, the random walk forecast errors fail to be orthogonal to the change in consumer opinion about gasoline prices, fuel efficiency and poor selections (quality). Such findings remain unchanged when we use the ‘weakly’ rational forecasts of gasoline prices instead of the random walk benchmark. Put together, our results provide new insights into the potential usefulness of consumer vehicle‐buying assessments for predicting gasoline prices. We recommend forecasters pay special attention to such assessments when forecasting retail gasoline prices, since they can potentially help produce forecasts that are superior to both the random walk and ‘weakly’ rational forecasts.

The Complementarity Between Man and Machine in Forecasting

This paper quantifies the amount of noise and bias in analysts’ forecast of corporate earnings at various horizons. We first show analyst forecasts outperform statistical forecasts at short-horizons, but underperform at longer horizons. We next decompose the relative accuracy of these forecasts into three components: (i) noise, (ii) bias and (iii) analysts’ information advantage over statistical forecasts. We find the information advantage is constant across forecasting horizons, while both noise and bias are increase linearly. We then show most existing models lack a mechanism to account for these facts. To generate such a mechanism, we consider a parsimonious variant of the model of Patton and Timmermann (2010) with a noisy cognitive default and show it quantitatively fits the data. The intuition underlying this model is that forecasters rely on their biased and noisy defaults more at longer horizons, as rational forecasts are less accurate. This model also quantitatively matches two non-targeted empirical relationships: (i) analyst disagreement increases with horizon and (ii) noise is an increasing function of volatility. Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

Noise in Expectations: Evidence from Analyst Forecasts

This paper quantifies the amount of noise and bias in analysts’ forecast of corporate earnings at various horizons. We first show analyst forecasts outperform statistical forecasts at short-horizons, but underperform at longer horizons. We next decompose the relative accuracy of these forecasts into three components: (i) noise, (ii) bias and (iii) analysts’ information advantage over statistical forecasts. We find the information advantage is constant across forecasting horizons, while both noise and bias are increase linearly. We then show most existing models lack a mechanism to account for these facts. To generate such a mechanism, we consider a parsimonious variant of the model of Patton and Timmermann (2010) with a noisy cognitive default and show it quantitatively fits the data. The intuition underlying this model is that forecasters rely on their biased and noisy defaults more at longer horizons, as rational forecasts are less accurate. This model also quantitatively matches two non-targeted empirical relationships: (i) analyst disagreement increases with horizon and (ii) noise is an increasing function of volatility.

Improved Water Quality Prediction with Hybrid Wavelet-Genetic Programming Model and Shannon Entropy

Prediction of biochemical oxygen demand (BOD) as the main pollution indicators of organic pollution in freshwater resources is necessary. In the present work, a hybrid wavelet-genetic programming (WGP) method was implemented to improve prediction of BOD. The Shannon entropy was used to identify the optimal input combinations of WGP. In addition, an investigation was done to find which functions of wavelet and decomposition levels have better results in conjunction with genetic programming (GP). For comparison of WGP efficiency, five machine learning methods consisting of WANN (wavelet-artificial neural network), ANN (artificial neural network), GP, DT (decision tree) and BN (Bayesian network) were considered. Experiments on wavelet-processed data revealed that the best results were obtained when the models WGP and WANN were calibrated at three levels of decomposition using the Dmey mother wavelet function. The WGP model created rational forecasts for the peak BOD values. The results show that the use of Shannon entropy is suitable for determining the optimal composition of inputs to machine learning methods. Comparison of the results indicate that the WGP model is superior to the GP, ANN, DT, BN and WANN models based on data from the Varian Hotel and Dam Input stations.

Do excessively volatile forecasts impact investors?

There is a logical bound on the time-series variability of analyst forecasts; when variability exceeds this bound it must be caused by something besides statistically rational forecasting. We document occurrences of excessively volatile analyst forecasts and show that they influence investment performance. Comparing trading rules based on forecasts that are excessively volatile and those that are not, we find the returns to investing based on the former are significantly lower, with higher daily volatility, and a lower Sharpe ratio. We also show that returns to trading based on excessively volatile forecasts underperform the most when there is little news arriving and when the news that does arrive is relatively neutral. In this region, it is hardest to argue that analysts are unwittingly overreacting to news; instead, they appear to be intentionally making extreme forecasts to curry favor with management or to differentiate themselves from other analysts.

The Role of Future Economic Conditions in the Cross-Section of Stock Returns: Evidence From the US and UK

Our study investigates the explanatory power of future economic conditions on individual stock returns in the US and UK equity markets. We analyse a new trading strategy that is based on rational forecasts of future real activity. In addition, we specifically examine the performance of this trading strategy applied to two different classifications of stocks – procyclical stocks and countercyclical stocks. Our findings indicate a strong persistence in the relationship between returns on pro-cyclical stocks and the business cycle. However, such persistence is not present when moving to counter-cyclical stocks in the US and the UK. From this we suggest that US and UK equity investors who predict future real activity accurately can improve their investment profitability by longing pro-cyclical stocks when they expect future economic conditions to be above the long-run trend and shorting those stocks when future activity is anticipated to be below the steady state.

Inflation uncertainty and inflation expectations: Micro-level evidence from the eurozone

This paper investigates the evolution of consumers’ inflation expectation. Using Eurobarometer survey data, we estimate models of the probability and the size of “forecast failure” of inflation forecasts. We document that higher inflation uncertainty, higher lagged inflation rates and increased inflation news volume reduce both the probability of failing to predict the correct inflation trajectory and the probability of making large forecast errors – in line with models featuring rational forecasting and sticky information. Furthermore, when the volatility of shocks to inflation increases, individuals tend to make incorrect inflation forecasts as well as large forecast ‘mistakes’, consistent with models featuring asymmetric loss functions. We also document heterogeneity in expectation formation related to demographic characteristics of respondents such as age, gender or education.

Do Excessively Volatile Forecasts Impact Investors?

There is a logical bound on the time-series variability of analyst forecasts; when variability exceeds this bound it must be caused by something besides statistically rational forecasting. We document occurrences of excessively volatile analyst forecasts and show that they influence investment performance. Comparing trading rules based on forecasts that are excessively volatile and those that are not, we find the returns to investing based on the former are significantly lower, with higher daily volatility, and a lower Sharpe ratio. We also show that returns to trading based on excessively volatile forecasts underperform the most when there is little news arriving and when the news that does arrive is relatively neutral. In this region, it is hardest to argue that analysts are unwittingly overreacting to news; instead, they appear to be intentionally making extreme forecasts to curry favor with management or to differentiate themselves from other analysts.

Rational Inattention and Analyst Forecast Accuracy

Facing limited attention constraints, financial analysts must strategically choose which information to pay attention to and which information to ignore when making earnings forecasts. I rely on rational inattention theory to develop and test hypotheses on factors that determine analyst attention and how analyst attention affects forecast accuracy. I construct a novel measure of attention that varies across stocks followed by the same analyst during the same fiscal period. I find that analyst attention is decreasing in the marginal cost of paying attention and that analysts allocate attention to firms with earnings that are difficult to forecast, suggesting that analysts suffer from a limited attention constraint. Importantly, I find that attentive analysts are more accurate, and the effect of attention is larger for inexperienced analysts and stocks with highly volatile earnings. Finally, I find that attentive analysts are less likely to revise forecasts upwards and are less optimistic even after controlling for firm information environment and analyst ability.

Do markets learn to rationally expect US interest rates? An anchoring approach

ABSTRACTWe propose an augmented and dynamic forecast anchoring model to examine whether a group of rational forecasters coexists with or emerges besides a group of forecasters employing heuristic rules. This model is consistent with the economically rational expectations theory. Using experts’ 3-month and 10-year Treasury bill rate survey expectations at short and long horizons, we find that aggregate expectations fail to exhibit a learning process towards rationality. While forecasters essentially anchor their judgements to heuristics, a small proportion of agents rationally forecast the short-term interest rate, possibly due to Federal Reserve’s transparency practice in the conduct of monetary policy and forward guidance at the zero lower bound.

Investor sentiment and the prediction of stock returns: a quantile regression approach

ABSTRACTWe employ quantile regression to provide a detailed picture of the stock return forecasting ability of investor sentiment. We find that investor sentiment predicts aggregate stock returns at lower quantiles. However, the forecasting power is lost at upper quantiles. The results are robust after controlling for a comprehensive set of macroeconomic and financial predictors and for characteristic portfolios. We also show that investor sentiment consists mainly of cash flow news and contains little information about discount rate news. The ability to forecast cash flows increases gradually from the lower quantiles to upper quantiles. Our results do not support that the ability of investor sentiment to predict stock returns comes from a rational forecast of future cash flows.

Fundamental factors and extrapolation in stock-market expectations: The central role of structural change

Rational expectations and behavioral-finance models are widely interpreted as representing two distinct conceptions of decision-making: rational and irrational, respectively. Using survey data, this paper presents econometric evidence that both fundamental factors and extrapolation drive participants’ expectations of stock returns, but that they do so in ways that vary over time. Although both the REH and behavioral-finance approaches offer relevant insights for understanding participants’ expectations, neither of these distinct model classes is consistent with time-series data. The paper’s findings also suggest that structural change gives rise to ambiguity about the correct quantitative model driving outcomes. This ambiguity, faced by economists and market participants alike, is the key to according both fundamental and behavioral factors a role in rational forecasting.

History of antibiotic adaptation influences microbial evolutionary dynamics during subsequent treatment.

Antibiotic regimens often include the sequential changing of drugs to limit the development and evolution of resistance of bacterial pathogens. It remains unclear how history of adaptation to one antibiotic can influence the resistance profiles when bacteria subsequently adapt to a different antibiotic. Here, we experimentally evolved Pseudomonas aeruginosa to six 2-drug sequences. We observed drug order–specific effects, whereby adaptation to the first drug can limit the rate of subsequent adaptation to the second drug, adaptation to the second drug can restore susceptibility to the first drug, or final resistance levels depend on the order of the 2-drug sequence. These findings demonstrate how resistance not only depends on the current drug regimen but also the history of past regimens. These order-specific effects may allow for rational forecasting of the evolutionary dynamics of bacteria given knowledge of past adaptations and provide support for the need to consider the history of past drug exposure when designing strategies to mitigate resistance and combat bacterial infections.

What Calls to ARMs? International Evidence on Interest Rates and the Choice of Adjustable-Rate Mortgages

The relative popularity of adjustable-rate mortgages (ARMs) and fixed-rate mortgages (FRMs) varies considerably both across countries and over time. We ask how movements in current and expected future interest rates affect the share of ARMs in total mortgage issuance. Using a nine-country panel and instrumental variables methods, we present evidence that near-term (one-year) rational forecasts of future movements in ARM rates do affect mortgage choice, particularly in more recent data since 2001. However, longer-term (three-year) rational forecasts of ARM rates have a relatively weak effect, and the current spread between FRM and ARM rates also matters, suggesting that households are concerned with current interest costs as well as with lifetime cost minimization. These conclusions are robust to alternative (adaptive and survey-based) models of household expectations. Data, the online appendix, and the institutional appendix are available at https://doi.org/10.1287/mnsc.2016.2629 . This paper was accepted by Amit Seru, finance.

A State Space Approach to Evaluate Multi-Horizon Forecasts

We propose a state space modeling framework to evaluate a set of forecasts that target the same variable but are updated along the forecast horizon. The approach decomposes forecast errors into three distinct horizon-specific processes, namely, bias, rational error and implicit error, and attributes forecast revisions to corrections for these forecast errors. We derive the conditions under which forecasts that contain error that is irrelevant to the target can still present the second moment bounds of rational forecasts. By evaluating multi-horizon daily maximum temperature forecasts for Melbourne, Australia, we demonstrate how this modeling framework analyzes the dynamics of the forecast revision structure across horizons. Understanding forecast revisions is critical for weather forecast users to determine the optimal timing for their planning decision.

Anticipated Banking Panics

We develop a macroeconomic model with banking instability. Sunspot runs can arise that are harmful to the economy. However, whether a run equilibrium exists depends on fundamentals. In contrast to earlier work, the probability of a sunspot run is the outcome of rational forecast based on fundamentals. The model captures the movement from slow to fast runs that was a feature of the Great Recession: A weakening of banks' balance sheets increases the probability of a run, leading depositors to withdraw funds from banks. These slow runs have harmful effects on the economy and set the stage for fast runs.