Ingmar Nolte

Improved Inference in Regression with Overlapping Observations

We present an improved method for inference in linear regressions with overlapping observations. By aggregating the matrix of explanatory variables in a simple way, our method transforms the original regression into an equivalent representation in which the dependent variables are non-overlapping. This transformation removes that part of the autocorrelation in the error terms which is induced by the overlapping scheme. Our method can easily be applied within standard software packages since conventional inference procedures (OLS-, White-, Newey-West- standard errors) are asymptotically valid when applied to the transformed regression. Through Monte Carlo analysis we show that it performs better in finite samples than the methods applied to the original regression that are in common usage. We illustrate the significance of our method with three empirical applications.

Improved inference and estimation in regression with overlapping observations

We present an improved method for inference in linear regressions with overlapping observations. By aggregating the matrix of explanatory variables in a simple way, our method transforms the original regression into an equivalent representation in which the dependent variables are non-overlapping. This transformation removes that part of the autocorrelation in the error terms which is induced by the overlapping scheme. Our method can easily be applied within standard software packages since conventional inference procedures (OLS-, White-, Newey-West-standard errors) are asymptotically valid when applied to the transformed regression. Through Monte Carlo analysis we show that it performs better in finite samples than the methods applied to the original regression that are in common usage. We illustrate the significance of our method with three empirical applications.

How Do Individual Investors Trade

This paper examines how high-frequency trading decisions of individual investors are influenced by past price changes. Specifically, we address the question as to whether decisions to open or close a position are different when investors already hold a position compared with when they do not. Based on a unique data set from an electronic foreign exchange trading platform, OANDA FXTrade, we find that investors’ future order flow is (significantly) driven by past price movements and that these predictive patterns last up to several hours. This observation clearly shows that for high-frequency trading, investors rely on previous price movements in making future investment decisions. We provide clear evidence that market and limit orders flows are much more predictable if those orders are submitted to close an existing position than if they are used to open one. We interpret this finding as evidence for the existence of a monitoring effect, which has implications for theoretical market microstructure models and behavioral finance phenomena, such as the endowment effect.

A detailed investigation of the disposition effect and individual trading behavior: a panel survival approach

This article uses a panel survival approach to analyze the trading behavior of foreign exchange traders. We concentrate on a detailed characterization of the shape of the disposition effect over the entire profit and loss regions. In doing so, we investigate the influence of a number of trading characteristics on the impact of the disposition effect. These trading characteristics include: special limit order strategies, trading success, size and the experience of our investors. Our main findings are that (i) the disposition effect has a nonlinear shape. For small profits and losses we find an inverted disposition effect, while for larger ones, the usual positive disposition effect emerges. (ii) The inverted disposition effect is driven to a great extend by patient and cautious investors closing their positions with special limit orders (take-profit and stop-loss). The normal positive disposition effect is found to be intensified for impatient investors closing their positions actively with market orders. (iii) We show that unsuccessful investors reveal a stronger inverse disposition effect. (iv) Evidence that bigger investors are less prone to the disposition effect than smaller investors is also found.

Estimating High-Frequency Based (Co-) Variances: A Unified Approach

We propose a unified framework for estimating integrated variances and covariances based on simple OLS regressions, allowing for a general market microstructure noise specification. We show that our estimators can outperform, in terms of the root mean squared error criterion, the most recent and commonly applied estimators, such as the realized kernels of Barndorff-Nielsen, Hansen, Lunde & Shephard (2006), the two-scales realized variance of Zhang, Mykland & Ait-Sahalia (2005), the Hayashi & Yoshida (2005) covariance estimator, and the realized variance and covariance with the optimal sampling frequency derived in Bandi & Russell (2005a) and Bandi & Russell (2005b). For a realistic trading scenario, the efficiency gains resulting from our approach are in the range of 35% to 50%.

Customer Trading in the Foreign Exchange Market Empirical Evidence from an Internet Trading Platform

This paper analyzes the relationship between currency price changes and their expectations. Currency price change expectations are derived with the help of different order flow measures, from the trading behavior of investors on OANDA FXTrade, which is an internet trading platform in the foreign exchange market. We investigate whether forecasts of intra-day price changes on different sampling frequencies can be improved with the information contained in the flow of our investors’ orders. Moreover, we verify several hypotheses on the trading behavior and the preference structure of our investors by investigating how past price changes affect future order flow.

Disagreement versus uncertainty: Evidence from distribution forecasts.

We use a cross-section of economic survey forecasts to predict the distribution of US macro variables in real time. This generalizes the existing literature, which uses disagreement (i.e., the cross-sectional variance of survey forecasts) to predict uncertainty (i.e., the conditional variance of future macroeconomic quantities). Our results show that cross-sectional information can be helpful for distribution forecasting, but this information needs to be modeled in a statistically efficient way in order to avoid overfitting. A simple one-parameter model which exploits time variation in the cross-section of survey point forecasts is found to perform well in practice.This paper generalizes the discussion about disagreement versus uncertainty in macroeconomic survey data by emphasizing the importance of the (unknown) true predictive density. Using a forecast combination approach, we ask whether cross sections of survey point forecasts help to approximate the true predictive density. We find that although these cross-sections perform poorly individually, their inclusion into combined predictive densities can significantly improve upon densities relying solely on time series information.

Sell-Side Analysts' Career Concerns during Banking Stresses

We propose a new approach to examine sell-side analysts’ career concerns by relating their forecast boldness to their employers’ news flows. Specifically, we use banking sector news to proxy for the severity of career concerns. Analysts follow more closely the consensus forecast when the prospects of the banking sector are negative (and vice versa). The effect is both economically and statistically significant after controlling for various firm, analyst, brokerage house, and forecasting characteristics, as well as sector and economy wide effects. The more established analysts, in terms of reputation and experience, are generally unaffected by banking sector news. In contrast, their less established peers tend to cluster their forecasts near the consensus after a sequence of negative news flows for banks. Collectively, our results support the notion that during banking stresses when job security is low analysts’ tendency to imitate others increases.

Where Do the Joneses Go on Vacation? Social Distance and the Influence of Online Reviews on Product Sales

This paper provides field evidence that social distance between customers and reviewers influences the impact from online reviews on product sales. We conceptualize information on interpersonal similarity as a heuristic cue that helps customers to infer similarity in product preferences between reviewers and customers. We hypothesize that customers put less weight on information from socially distant others. We test this hypothesis on an extensive dataset of 60,000 hotel-week observations from a large online travel and holiday portal. The data allows for measuring interpersonal similarity in a clean way, because reviewers need to specify whether they traveled as couples, families, or singles, and reviews can be traced back to corresponding bookings. Therefore, we know for each reviewer, if at the time of booking her peers were couples, families, or singles. Our results provide strong empirical support for our hypothesis. For example, singles increase hotel demand by 18% in response to positive peer information, but discard family information. We conclude that firms’ management of online reviews must reflect on social distance between reviewers and target customers.

A Multivariate Integer Count Hurdle Model: Theory and Application to Exchange Rate Dynamics

In this paper we propose a model for the conditional multivariate density of integer count variables defined on the set Zn. Applying the concept of copula functions, we allow for a general form of dependence between the marginal processes which is able to pick up the complex nonlinear dynamics of multivariate financial time series at high frequencies. We use the model to estimate the conditional bivariate density of the high frequency changes of the EUR/GBP and the EUR/USD exchange rates.