Toward Better Risk Forecasts

Properties of range-based volatility estimators

This research investigates the role of trading volume and data frequency in volatility forecasting by evaluating the performance of Generalized Autoregressive Conditional Heteroskedasticity Mixed‐Data Sampling (GARCH‐MIDAS), traditional GARCH, and intraday GARCH models. We take trading volume as the proxy for information flow and examine whether the Sequential Information Arrival Hypothesis (SIAH) is supported in the China stock market. The contributions of this study are as follows. (1) We provide a more consistent comparison to evaluate the forecasting ability of the MIDAS approach. (2) We extend the literature on the forecasting performance of trading volume to the GARCH‐MIDAS approach. (3) We present clear evidence to support that forecasting ability strongly relies upon data frequency. The empirical results show that: (1) GARCH‐MIDAS is not able to beat the traditional GARCH method when both are estimated by the same predictor sampled at different frequencies; (2) there is a positive relation between trading volume and volatility, but no clear evidence appears that SIAH holds in the China stock market; and (3) high‐frequency data are highly recommended for daily realized volatility (RV) forecasting, whereas intraday GARCH could significantly outperform traditional GARCH and GARCH‐MIDAS in volatility forecasting.

The Directional Change (DC) is an event approach for summarizing price movements in financial markets based on a given threshold value. The Dynamic Threshold Trading Strategy (DT-TS) is a trading strategy built based on the DC approach with daily dynamic defined threshold values. The DT-TS trading action is triggered, if a DC event is detected (price change is spotted) and the prices continue to increase or decrease. The Buy or Sell trading action to be taken depends on the previous day price changes (short term history). In this paper, we apply the DT-TS trading strategy on data streams of different frequency to investigate its potential to generate surplus. One major issue under examination is the strategy's applicability and ability to provide good results on data streams of different frequency (higher versus lower frequency). An experiment was conducted using the FTSE 100 index min-bymin and day-by-day streams for more than 35 weeks. Results showed that the DT-TS works better (gained higher profits) with higher frequency data rather than lower frequency data.

Operative Use of Eddy Covariance Measurements: Are High Frequency Data Indispensable?

An acoustic field is governed by the equation Del 2u+ omega 2u+ omega 2a1(x)u+ Del .(a2(x) Del u)=- delta (x-y) in R3 and u is measured on the surface of the Earth, i.e. on the plane x3=0 for all positions of the source y and receiver x and for all frequencies. The authors show that the low frequency portion of the data determines a2(x) while the high frequency portion of the data determines (a1-a2)(1+a2)-1 identical to p(x). Here a2= rho -1 Delta rho is the relative variation of the density of the inhomogeneity and p(x) depends only on the relative variation of the velocity in the inhomogeneity. (ii) They show how to recover a2 from the low frequency data and how to recover p(x) from the high frequency data. They give a formula for the high frequency asymptotics of u on the plane x3=0.

Mapping reservoir architecture (geobody size, shape, and stacking patterns) in the subsurface is critical for exploring and producing hydrocarbons, CO2 storage, and geothermal resource development since it can define connectivity or compartmentalization of flow zones (Meirovitz et al. 2020). However, our capacity to interpret depositional system architecture is limited by seismic resolution. In addition to limited bandwidth, the resolution of discrete geologic features mapped by seismic attributes can be mixed through the vertical analysis window. In this work, we use synthetic seismic data derived from an outcrop analogue to better understand how seismic bandwidth affects the vertical and areal resolution of stacked stratigraphic features. We studied five synthetic seismic volumes from low to high-frequency bandwidths of 15 Hz, 30 Hz, 60 Hz, 90 Hz, and 180 Hz from a deepwater channelized slope system in the Magallanes Basin, Chile. We analyze the effect of different seismic attributes: coherence, dip magnitude, dip azimuth, root mean square amplitude, and Laplacian filters on our different bandwidth data to understand how much “mixing” of stratigraphic features there is by comparing with the true geological model. We explore how the attributes’ parametrization affects the imaging of differently sized features by modifying the analysis window in each case from +/-2ms to +/- 100 ms. Results show that the “mixing” occurs as a result of 1) the seismic bandwidth, 2) the algorithm used for each seismic attribute calculation, and 3) the attribute analysis window. Broad band, higher frequency data and small analysis windows provide clear images of the stacked channels. In contrast, low-frequency data and larger analysis windows result in more mixing or “composite” appearance, affecting interpretations and NTG estimates. The Laplacian filter’s use proves to enhance the distinction of the different channel architectures providing high-resolution edge detection even for the lower frequency data.

Consistent High-Precision Volatility from High-Frequency Data

Dynamic Copula Models and High Frequency Data

Floating car data (FCD) application is an urban sensing application, which collects the status (position, velocity and heading) of vehicles to feed a traffic management server and estimate the travel time of vehicles, and can be carried by long term evolution (LTE) or a heterogeneous network composed by LTE and vehicular ad hoc network (VANET). However, this application is characterized by high frequency and small data size, which occupy network resource frequently and may have an impact on other LTE traffic, such as human to human (H2H) traffic, which is characterized by low frequency and big data size. In this paper, we aim to analyze and reduce the impact of FCD traffic on the H2H traffic in a LTE-VANET heterogeneous network. For this purpose, we design a novel FCD transmission scheme and evaluate its impact on H2H traffic, which is represented by the rise of the blocking ratio of the H2H traffic, through a Markovian model. The results of the model show that comparing with other schemes, the novel scheme has a relatively lower impact on the traditional cellular traffic, H2H traffic.

Climate Change is a global challenge and needs to be addressed immediately. The emission of Green House Gases in the atmosphere by anthropogenic factors is one of the major causes of Global Warming and Climate Change. India is working towards the control of global warming by focussing on controlling Green House Gas emissions. The emission of carbon-di-oxide in the atmosphere plays a predominant role in global warming. The Global Warming potential of the Green House Gases is measured in terms of carbon-dioxide equivalent CO2 (eq.). The annual CO2 (eq.) emission data from 2005-2015 from four key sectors viz., Energy, Industry Process and Product Use (IPPU), Agriculture Forestry and Land Use (AFOLU), and the Waste sector were considered in the study. Classical Temporal disaggregation methods Denton, Denton Cholette, Chow - Lin, Fernandez, and Litterman methods were employed to disaggregate the low frequency (annual) data to high frequency (quarterly) data. The analysis revealed that the Chow - Lin method of disaggregation best suited to disaggregate the CO2 (eq.) series for the three sectors except AFOLU with anAdjusted R square of 0.9 and the current Price GDP is the good indicator series for CO2 (eq.). The disaggregated data is modelled using ARIMA modelling. The CO2 (eq.) from 2021-Q1 to 2023-Q4 is forecasted using the fitted ARIMA model for each sector.

The last decade has seen substantial advances in the measurement, modelling and forecasting of volatility which has centered around the realized volatility literature. To date, most of the focus has been on the daily and monthly frequencies, with little attention on longer horizons such as the quarterly frequency. In finance applications, forecasts of volatility at horizons such as quarterly are of fundamental importance to asset pricing and risk management. In this article we evaluate models for stock return volatility forecasting at the quarterly frequency. We find that an autoregressive model with one lag of quarterly realized volatility with an in-sample estimation period of between 60 and 80 quarters produces the most accurate forecasts, and dominates other approaches, such as the recently proposed mixed-data sampling (MIDAS) approach.

Missing data is a common problem encountered by empirical researchers and practitioners. This dissertation is a collection of three essays on handling imperfectly observed economic data. The first essay addresses temporal aggregation where some high frequency data are missing but their sum or average are observed in the form of low frequency data. In a vector autoregression model with varied frequency data, the explicit form of the likelihood function and the posterior distribution of missing values are found without resorting to the recursive Kalman filter. The second essay further discusses data aggregation in a two-equation model in which the missing values are imputed by a regression. In two scenarios, the likelihood function is shown to be separable and the analytic maximum likelihood estimator can be obtained by two auxiliary regressions, which is advantageous to the conventional least squares imputation approach in terms of both efficiency and computability. The third essay concerns the finite-sample bias of estimators associated with the monotone instrumental variables, which is a useful assumption to partially identify the counterfactual outcomes. It is shown that a multi-level bootstrap procedure can reduce and gradually eliminate the bias. A simultaneous simulation strategy is also proposed to make multi-level bootstrap computationally feasible.

The degree of variation of trading prices with respect to time is volatility-measured by the standard deviation of returns. We present the estimation of stochastic volatility from the stochastic differential equation for evenly spaced data. We indicate that, the price process is driven by a semi-martingale and the data are evenly spaced. The results of Malliavin and Mancino [1] are extended by adding a compensated poisson jump that uses a quadratic variation to calculate volatility. The volatility is computed from a daily data without assuming its functional form. Our result is well suited for financial market applications and in particular the analysis of high frequency data for the computation of volatility.

Financial prices are often discretized—with smallest tick size of one cent, for example. Thus prices involve rounding errors. Rounding errors affect the estimation of volatility, and understanding them is critical, particularly when using high frequency data. We study the asymptotic behavior of realized volatility (RV), which is commonly used as an estimator of integrated volatility. We prove the convergence of the RV and scaled RV under varous conditions on the rounding level and the number of observations. A bias-corrected volatility estimator is proposed and an associated central limit theorem is shown. The simulation and empirical results demonstrate that the proposed method can yield substantial statistical improvement.

The construction and functioning principles of two-channel equipment for simultaneous acoustic profiling with two different frequency sources are described. The equipment makes possible the collection of high resolution data for the upper part of the sediments and reasonable depth penetration for engineering purposes. Unlike profiling by applying two independent systems, the presented hard- and software package makes it possible to synchronize the sources and eliminate mutual interference. The data is saved to a common file on a single computer. The software package contains resources for processing the high and low frequency data.