Variance Ratio Test Research Articles

Estimation of battery state of charge based on changing window adaptive extended Kalman filtering

Adaptive extended Kalman filter (AEKF) is commonly used for lithium-ion battery state of charge (SOC) estimation. However, it overlooks the impact of changes in the distribution of error innovation sequence (EIS) on the noise covariance, resulting in inaccurate state of charge estimates. To address this issue, this paper introduces a novel approach called changing window adaptive extended Kalman filter (CW-AEKF) algorithm. This algorithm uses variance ratio and Levene test to identify the change in the distribution of error innovation sequence, and adaptively updates the optimal noise window length based on the judgment result to achieve accurate noise estimation. Subsequently, the proposed algorithm is combined with a temperature-corrected second-order RC equivalent circuit model for state of charge estimation. The results of dynamic stress test (DST) at different temperatures show that the changing window adaptive extended Kalman filter algorithm can obtain higher accuracy in state of charge estimation results than other algorithms, with state of charge estimation errors remaining within 1 %. Finally, the state of charge estimation of the changing window adaptive extended Kalman filter algorithm in publicly available datasets is analyzed. The results demonstrate that the proposed algorithm maintains strong generalization ability when facing various working conditions.

Adaptive Market Efficiency Hypothesis in ASEAN Stock Markets: A Variance Ratio Analysis Across Market Capitalization Segments

This study examines the efficiency of the ASEAN stock market and investigates the dynamic adaptability of this efficiency over time. Utilizing advanced methodologies, including the Multiple Variance Ratio (MV) and Wild Bootstrapped Variance Ratio (WBVR) tests, in conjunction with the Rolling Window technique, the research assesses market efficiency across different periods from March 2009 to March 2024. The analysis is conducted using daily price data, segmented into the overall market and three sub-groups based on market capitalization, to explore the potential impact of company size on market efficiency. The findings indicate that market efficiency is not static; it fluctuates, particularly during periods of economic crises or significant events, when all market segments exhibit inefficiency, deviating from the Random Walk theory. This suggests that during such periods, stock prices become more predictable, contrary to the expectations of an efficient market. Additionally, the study finds that changes in market efficiency are consistent across different company sizes, suggesting that market or company size does not significantly influence efficiency.

Evolving Efficiency of Cryptocurrency Market: Evidence from Leading Cryptocurrencies

The article investigated the efficiency of the cryptocurrency market, with a focus on Bitcoin (BTC), Ethereum (ETH), Tether (USDT), and Binance Coin (BNB-USD), based on their substantial market capitalization. For this purpose, we observed the daily closing prices spanning from January 2018 to December 2023 and employed a set of robust tests, including the generalized spectral test, automatic portmanteau test, and automatic variance ratio test. The findings of the study reveal a random pattern in price fluctuations, indicating weak form efficiency. Furthermore, we adopted the rolling window approach to investigate whether market efficiency is dynamic or static over time. The empirical result illustrated that the crypto market efficiency remains static over time, except for USDT. In conclusion, the overall empirical results support the notion of the random walk hypothesis, indicating that past price movements offer no predictive insight into future prices. These findings have significant implications for investors, emphasizing the lack of predictive insight from past price movements. Policymakers are urged to establish a robust framework for market integrity and reliable price discovery.

The Inefficiency of Bitcoin and the COVID-19 Pandemic

From the perspectives of asset pricing, market outreach, regulatory framework, and investor ethos, crypto markets differ substantially from traditional financial markets. Given these fundamental differences, it is interesting to examine how the notions of market efficiency apply to crypto markets, especially because the arguments of the efficient market hypothesis (Fama, 1970) and the adaptive market hypothesis (Lo, 2004, and Lo, 2008) were originally developed in the context of traditional financial markets. Research on the informational efficiency of crypto markets has attracted increasing attention in recent years. This paper examines the evolving efficiency of Bitcoin, the leading cryptocurrency, especially during the period when economies around the world were devastated by the Covid-19 pandemic. For a newly-emerged cryptocurrency with a market that is essentially global, a global shock like the Covid-19 pandemic presents ideal conditions for assessing how efficiency evolves as the market faces a series of shocks. Employing a fixed-length rolling window approach, this paper carries out the following tests: the automatic portmanteau test of Escanciano and Lobato (2009), the wild bootstrap automatic variance ratio test proposed by Kim (2009), the generalized spectral test of Escaciano and Valesco (2006), and the test proposed by Dominguez and Lobato (2003). The results provide evidence of episodes of inefficiency in a market that is efficient over extended periods. The inefficiency index constructed in this study shows that the Bitcoin market went through proportionately longer and more frequent episodes of inefficiency during the Covid-19 period. This is in line with the adaptive market hypothesis and has practical significance for investors and regulators.

Testing the Adaptive Market Hypothesis and Time-Varying Efficiency in the Indian Equity Market

The study examines the adaptive market hypothesis (AMH) as an evolutionary principle of the alternative efficient market hypothesis in the Indian stock market (Sensex and Nifty50) on the daily return from April 2014 to May 2020. Based on AMH, investors behave, learn, and adapt to market conditions. This distinction of dynamic market conditions is divided into bull and bear market classifications. We apply three variations of the variance ratio test and the returns have been whitened using the Autoregressive model with generalized autoregressive conditional heteroskedasticity (AR-GARCH) approach to examine the nonlinear predictability test. Further, we evaluate a fixed-length subsample window framework to detect the time-varying predictability and examine whether market conditions affect stock return predictability and market condition. The study confirms inefficient market behaviour during crises, fear, panic, macroeconomic events and each market adapts differently to certain market conditions. Furthermore, the return series exhibits significant periods of efficiency and inefficiency consistent with the adaptive market hypothesis. The evidence of our findings sheds light on efficient market behaviour in dynamic market conditions and market environments for researchers, investors in investment strategies and policy intervention in risk containment measures and control market manipulation.

Comparative analysis of stochastic seasonality, January effect and market efficiency between emerging and industrialized markets

This study investigates whether there are significant differences in investment returns between emerging markets and industrialized markets in terms of stochastic seasonality, January effect and market efficiency. Data on investments, and returns for nine emerging countries and eleven industrialized countries spanning January 1990 to December 2020, were obtained from the Organization for Economic Cooperation and Development (OECD). The spectral nonparametric test was used to determine the presence of stochastic seasonality for each market while the regression test was used to determine the presence of January effect. In the case of determining the efficiency status of the markets, the variance ratio test and the runs test were used. In cases where there appeared to be differences between the two types of market, Fisher's exact test was used to verify such differences. The results show no significant differences between the two types of markets in terms of seasonality, January effect and efficiency statuses. Apart from Brazil which recorded stochastic seasonality, all others are not stochastically seasonal. In the case of the January effect, it was a mixed bag; five emerging markets had January effect while two industrialized markets had January effect.

Random walk tests for the MENA stock returns

Purpose — The current study seeks to understand whether individual stock returns exhibit random movement and are not dependent (efficient at weak form) on fourteen out of sixteen actively traded Arab stock markets in the Middle East and North Africa (MENA) region, based on the size of the market value.Method — Various non-parametric methods, including autocorrelation test, variance ratio test, Phillips-Perron unit root test, and runs test, are used to assess the random walk hypothesis for daily data following the Covid-19 vaccination program. This analysis covers the period from January 3, 2021, to March 28, 2023.Result — The study results present evidence that all individual stock returns deviate from random walk behavior. However, only Kuwait, Jordan, and Palestine stock returns follow the random walk based on the run test results at a significance level of 10%. Therefore, it can be concluded that all stock returns are inefficient at the weak-form, suggesting that investors have opportunities for unexpected gains.Practical implications — The findings of this study suggest that investors in the MENA region may have opportunities for unexpected gains, as individual stock returns deviate from random walk behavior, highlighting the importance of considering market dynamics and employing informed investment strategies. Additionally, policymakers could benefit from understanding the inefficiencies in stock returns to implement measures that promote market stability and efficiency.

Commodity Market Efficiency: New Evidence From the Russia-Ukraine War

We re-evaluate the efficiency of selected commodity markets during the Russia-Ukraine crisis using the asymptotic normal variance ratio test. We find that there is a substantial difference in the commodity prices before and after the war’s commencement. We also show that the markets are weak-form inefficient before and after the invasion except for gasoline and oats which became efficient during the war. We highlight the investment implications of the findings.

Market sentiment and price dynamics in weak markets: A comprehensive empirical analysis of the soybean meal option market

AbstractThis study addresses key issues of market efficiency in weak global futures markets, focusing on the intricate relationship between market sentiment and options pricing. Employing rolling variance ratio tests and information‐sharing models for market dynamics analysis, and supplemented with Granger causality tests and impulse response findings, it reveals a significant, unidirectional impact of market sentiment on options pricing, especially during periods of heightened sentiment. These insights underscore the importance of considering time dynamics in market behavior analysis, offering a novel perspective on futures and options market understanding.

A residual‐based nonparametric variance ratio no‐cointegration test

It is prominently stated in the literature that local asymptotic power properties serve as a useful indicator for the performance of residual‐based no‐cointegration tests in finite samples. However, this article comes to an opposing conclusion. In particular, we show that Breitung's (2002, Journal of Econometrics 108, 343–363) nonparameteric variance ratio unit root test applied to regression residuals serves as a no‐cointegration test but is inferior to its competitors from a local asymptotic power perspective. Nevertheless, in finite samples, the variance ratio test has good size properties, competitive power, and the convenience of being tuning parameter free. In general, we find that short‐run dynamics in the error process can have considerably larger detrimental effects on the performance of residual‐based no‐cointegration tests in finite samples than changes in the only nuisance parameter affecting local asymptotic power of the tests. The results serve as a warning for practitioners and lead to interesting directions for future research.

Facial Surface Electromyography: A Novel Approach to Facial Nerve Functional Evaluation after Vestibular Schwannoma Surgery.

Vestibular schwannoma (VS) surgery may cause facial nerve damage. However, a comprehensive evaluation of post-operative facial outcomes may be difficult to achieve. Surface electromyography (sEMG) is a promising non-invasive evaluation tool. However, its use in the follow-up after VS surgery has not been reported yet. The main objective was to develop and validate a new sEMG application specifically for the post-VS surgery setting. Secondary goals were to provide a systematic description of facial muscle activity after VS surgery and assess the association between sEMG parameters and Sunnybrook scale scores. Thirty-three patients with facial palsy following VS surgery were included. The clinical outcomes (Sunnybrook symmetry, movement, and synkinesis scores) and sEMG parameters (signal amplitude normalized by the maximal voluntary contraction (NEMG) and sEMG synkinesis score (ESS, number of synkinesis per movement sequence)) were evaluated at the end of the follow-up. In all tested muscles, NEMG variance was significantly higher on the affected side than the contralateral (variance ratio test, p < 0.00001 for each muscle). In total, 30 out of 33 patients (90.9%) showed an ESS ≥ 1 (median: 2.5, IQR: 1.5-3.0). On the affected side, NEMG values positively correlated with both dynamic and overall Sunnybrook scores (Spearman's model, p < 0.05 for each muscle, except orbicularis oculi). ESS significantly correlated with the Sunnybrook synkinesis score (Spearman's rho: 0.8268, p < 0.0001). We described and preliminarily validated a novel multiparametric sEMG approach based on both signal amplitude and synkinesis evaluation specifically for oto-neurosurgery. Large-scale studies are mandatory to further characterize the semiological and prognostic value of facial sEMG.

Dynamic efficiency in MENA stock markets during COVID-19 outbreak and vaccines

PurposeThis paper examines the time-varying efficiency of nine major Middle East and North Africa (MENA) stock markets namely Egypt, Bahrain, UAE, Jordan, Saudi Arabia, Oman, Qatar, Morocco and Tunisia during times of COVID-19 pandemic outbreak and vaccines.Design/methodology/approachThe authors use two econometric approaches: (1) autocorrelation tests including the wild bootstrap automatic variance ratio test, the automatic portmanteau test and the Generalized spectral test, and (2) a non-Bayesian generalized least squares-based time-varying model with statistical inferences.FindingsThe results show that the degree of stock market efficiency of Egyptian, Bahraini, Saudi, Moroccan and Tunisian stock markets is influenced by the COVID-19 pandemic crisis. Furthermore, the authors find a tendency toward efficiency in most of the MENA markets after the announcement of the COVID-19's vaccine approval. Finally, the Jordanian, Omani, Qatari and UAE stock markets remain globally efficient during the three sub-periods of the COVID-19 pandemic outbreak.Originality/valueThe results have important implications for asset allocations and financial risk management. Portfolio managers may maximize the benefit of arbitrage opportunities by taking strategic long and short positions in these markets during downward trend periods. Policymakers should implement the action plans and reforms to protect the stock markets from global shocks and ensure the stability of the stock markets.

An Empirical Study on the Efficiency of China's Convertible Bond Market

Since 2017, with the increased demand for margin financing and securities lending in the capital market and the improvement of relevant regulations, China's convertible bond market has experienced a boom. As of March 9, 2024, China has issued a total of 823 convertible bonds, amounting to 1.3 trillion yuan. The stock of convertible bonds has expanded from 24 bonds totalling 55.9 billion yuan at the end of 2016 to 553 bonds totalling 853.3 billion yuan. The importance of convertible bonds in the capital market has become increasingly prominent, becoming an essential part of financial tool diversification and market development. However, academic research on the convertible bond market is relatively scarce, mostly focusing on pricing mechanisms, financing effects, and the impact on the underlying stock prices, with little attention paid to the market itself. This paper uses runs tests, autocorrelation tests, and variance ratio tests to analyze the efficiency of China's convertible bond market. By examining whether the daily returns of the convertible bond index follow a random walk, the paper tests if the market is weak-form efficient. The results show that China's convertible bond market has not reached weak-form efficiency. Subsequently, a new index sample was created from the top ten convertible bonds by trading volume in May 2024, and its daily returns were tested for adherence to the random walk model. The results were compared to the overall market's convertible bond index returns in the same period of 2023 to identify factors influencing market efficiency, providing directions and insights for future research.

Do stock prices follow random walk over day and night? -– evidence from Chinese stock market

ABSTRACT In this study, we decompose daily stock return into day and night returns and examine whether stock prices follow random walk over day and night in the Chinese stock market. Both variance ratio test and economic test reject the random walk hypothesis. Specifically, we find a significant return reversal from night to day but a significant return momentum from day to night. We also show that the unique T + 1 trading rule in China attenuates the return reversal from night to day but strengthens the return momentum from day to night.

Understanding the Match: The Effect of Signaling, Demographics, and Applicant Characteristics on Match Success in the Orthopaedic Residency Application Process.

As of the 2022 to 2023 match cycle, orthopaedic residency programs began offering applicants 30 signals as part of a preference signaling program. Many have assumed that signals would become powerful tools in the match process, yet no objective data currently exist analyzing their effect. This study aims to provide such analysis while also offering comparisons with other factors affecting match success. Self-reported survey data on applicants and applications from 2017 to 2023 from the Texas Seeking Transparency in Application to Residency database were queried. Variables associated with receiving interviews and match success were analyzed using two-sided Student t -tests, chi-squared tests, variance ratio testing, and receiver operating characteristic analysis. Compared with 2017 to 2022, 2023 applicants submitted fewer applications (61.8 versus 78; P < 0.001), received fewer interview offers (11.6 versus 13.8; P < 0.001), and interview offers were spread more evenly among applicants (SD, 6.82 versus 9.10; P < 0.001). For 2023 applications, odds of securing an interview were increased most by away rotations (odds ratios [OR] 61.8; P < 0.001), use of a signal (OR, 9.61; P < 0.001), and geographic connection (OR, 4.70; P < 0.001). Female applicants received more interview offers from signaled programs than their male counterparts (11.2 versus 8.94; P = 0.003). Applicant variables most predictive of match success in 2023 were interview offers (area under the receiver operating characteristic curve [AUC] = 0.802), step 2 CK score (AUC = 0.673), and step 1 score (AUC = 0.648). The preference signaling program seems to be accomplishing its goals of reducing applications and more evenly distributing interviews. Away rotations, signals, and geographic connections represent the strongest predictors of applications resulting in a successful match. Applicants must use their signals carefully to maximize their chance of success. Level III.

TESTING WEAK-FORM EFFICIENCY FOR THE SAUDI PARALLEL MARKET (NOMU)

Market efficiency is essential for a thriving stock exchange and has a substantial impact on capital investments, particularly in emerging markets. This article examines the weak-form efficiency of Nomu, the Saudi Arabian parallel stock exchange. The study employed various statistical tests on the collected closing values of the Nomu index from May 1, 2019, to May 31, 2023. The return normality and stationarity were investigated using the Jarque-Bera, Augmented Dickey-Fuller (ADF), and Phillips-Perron (PP) tests. The Variance Ratio (VR) test was used to assess autocorrelation, while the Wald-Wolfowitz Runs (WWR) test was utilized to examine random walk behavior. While the Runs test indicated some randomness, all other tests rejected their null hypotheses, indicating that the Nomu market is inefficient according to the weak-form model. This inefficiency suggests that historical trading data may predict future prices, which could discourage foreign capital inflows, thereby limiting Nomu's growth and investment appeal. The significance of these findings for regulators and policymakers in formulating market regulations and policies aimed at enhancing Nomu's efficiency and investor appeal cannot be overstated. This paper, with its robust methodology, is believed to be the first to evaluate the Nomu market's weak-form efficiency since its inception, contributing to a better understanding of market dynamics in emerging stock exchanges.

Verification of the Weak-Form Informational Efficiency of Fuel Markets in the Visegrad Group

The paper aims at examining the weak-form informational efficiency of fuel markets in the Visegrad Group (V4) countries: the Czech Republic, Hungary, Poland and Slovakia from January 2016 through December 2020. For this purpose, the following statistical tests were applied: the runs test, the variance ratio test, the autocorrelation tests, the unit root tests. The tests provided mixed results not giving a definitive answer to the question of whether V4 fuel markets were informationally efficient in a weak form. The only exception is Slovakia where gasoline and diesel prices followed random walk, providing evidence in favor of the weak-form informational efficiency of the market.

TESTING THE MARKET EFFICIENCY WHEN INTEREST RATES CHANGE: CASE IN INDONESIA

Until the end of September 2023, the interest rate policy in the Republic of Indonesia shifted from 3.50% to 5.75%. The objective of this study is to examine whether changes in interest rates have an impact on the market efficiency in Indonesia. The daily market returns of Indonesia Stock Exchange (IDX) indices from January 3rd, 2022, until October 9th, 2023 are employed. To examine the market efficiency in weak form, this study conducts the normality test by the Kolmogorov-Smirnov test, the unit root problem test by Augmented Dickey-Fuller test, and the variance ratio test. This study finds that most market indices experienced a decline in returns and were less volatile when interest rates shifted to higher levels. This study also finds that most market indices tend to remain efficient in weak form at both low and high interest rates. The finding implies that those indices tend not to be influenced by policy changes in interest rates. However, the findings also show that several market indices experience changes in conditions when interest rates shift. As the studies on the relationship between changes in interest rates and market efficiency are scarce, the novelty of this study is the use of event windows to get specific impact of changes in interest rates on market efficiency.

Impact of Tax Implementation on Fintech P2P Loans in Indonesia

This research paper aims to examine the impact of tax implementation on the number of P2P loan recipients and the amount of P2P loan disbursement in fintech industry of Indonesia. This empirical study is based on the number of accounts and loan data acquired from one of Indonesia’s government agencies – the Financial Services Authority (OJK) from February 2022 to April 2022 and June 2022 to August 2022 that are analyzed using various tests such as Skewness-Kurtosis Test, the variance ratio test, and t-test. The results show that tax implementation does not significantly affect the number of P2P loan recipients and the amount of P2P loan disbursement. This is because fintech offers greater benefits than traditional banks, such as convenience, time-saving, more access to financial products, and better security measures. The limited amount of available data and the number of provinces available become the limitation of the study to obtain optimum results.

Does the random walk hypothesis hold for Indian stock markets during parliamentary elections?

In this study, we examine the random walk hypothesis for two well-known Indian indices, BSE (Sensex) and NSE (Nifty), by considering country-specific political events (parliamentary elections). Two pertaining questions were studied. First, efficiency with respect to weak form, semi-strong form, and strong form; and second, the random walk pattern of the return by applying the new variance ratio tests. We use 21 years of daily closing stock price data of both the National Stock Exchange (NSE) Nifty Index and the Bombay Stock Exchange (BSE) Sensex Index. The hypothesis is tested by using both conventional and new variance ratio tests: The Lo–MacKinlay variance ratio test, the Chow–Denning test, and Wright’s Rank and Sign test. All three tests report that the return does not follow the random walk for the full sample, suggesting the possibility of making gains by exploiting various investing strategies. It was found that both Indian indices follow the random walk hypothesis during the phase of parliamentary elections. This study contributes to the existing literature on the Efficient Market Hypothesis (EMH).