scholarly journals The Month-of-the-Year Effect in the European, American, Australian and Asian Markets

Economies ◽  
2021 ◽  
Vol 9 (4) ◽  
pp. 168
Author(s):  
Gualter Couto ◽  
Pedro Pimentel ◽  
Catarina Barbosa ◽  
Rui Alexandre Castanho
Keyword(s):  
Linear Regression ◽  
Statistical Significance ◽  
Statistical Evidence ◽  
Excess Return ◽  
January Effect ◽  
European American ◽  
Average Return ◽  
Year Effect ◽  
Halloween Effect

This paper examines the existence of the month-of-the-year effects in four different continents, namely Europe, Asia, America, and Oceania. Nine indexes were analyzed in order to verify differences between monthly returns from January 1990 to December 2013, followed by an examination of the January effect, Halloween effect, and the October effect, testing for statistical significance using an OLS linear regression in order to verify whether those effects offer consistent opportunities for investors. Investors with globally diversified portfolios benefit from the Halloween effect, with a 1.2% average monthly excess return in winter and spring, while the pre-dotcom-bubble period had a better performance than the post-dotcom-bubble period. In the global post-dotcom-bubble period, there is statistical evidence for 1.60% and 1% lower average monthly returns in January (the January effect) and in months other than October (the October effect), respectively, contradicting the literature. The dotcom bubble seems to be responsible for the January effect differing from what might otherwise have been expected in the later period. There is no consistent and clear impact on continental incidence. The Halloween effect is revealed to be a fruitful strategy in the FTSE, DAX, Dow Jones, BOVESPA, and N225 indexes taken one-by-one. The January effect excess average return was only statistically significative for the pre-dotcom-bubble period for globally diversified portfolios. This paper contributes to a wider global and comparable view upon month-of-the-year effect.

scholarly journals Calendar anomalies in the Ukrainian stock market

2017 ◽  
Vol 14 (1) ◽  
pp. 104-114 ◽  
Author(s):  
Guglielmo Maria Caporale ◽  
Alex Plastun
Keyword(s):  
Stock Market ◽  
Simulation Analysis ◽  
January Effect ◽  
Year Effect ◽  
Average Analysis ◽  
Day Of The Week ◽  
Calendar Anomalies ◽  
Student’S T ◽  
Trading Simulation ◽  
Halloween Effect

This paper is a comprehensive investigation of calendar anomalies in the Ukrainian stock market. It employs various statistical techniques (average analysis, Student’s t-test, ANOVA, the Kruskal-Wallis test, and regression analysis with dummy variables) and a trading simulation approach to test for the presence of the following anomalies: day-of-the-week effect; turn-of-the-month effect; turn-of-the-year effect; month-of-the-year effect; January effect; holiday effect; Halloween effect. The results suggest that in general calendar anomalies are not present in the Ukrainian stock market, but there are a few exceptions, i.e. the turn-of-the-year and Halloween effect for the PFTS index, and the month-of-the-year effect for UX futures. However, the trading simulation analysis shows that only trading strategies based on the turn-of-the-year effect for the PFTS index and the month-of-the-year effect for the UX futures can generate exploitable profit opportunities that can be interpreted as evidence against market efficiency.

Islamic calendar and stock market behaviour in India

2018 ◽  
Vol 45 (11) ◽  
pp. 1550-1566
Author(s):  
Dharani Munusamy
Keyword(s):  
Stock Market ◽  
Least Square Method ◽  
Least Square ◽  
Stock Market Returns ◽  
Content Type ◽  
Ordinary Least Square ◽  
Average Return ◽  
Year Effect ◽  
The Common ◽  
The Impact

Purpose The purpose of this paper is to examine the behavior of the stock market returns in the different days of the week and different months of the year in accordance with the Islamic calendar. Further, the study estimates the risk-adjusted returns to test the performance of the indices during the Ramadan and non-Ramadan days. Finally, the study investigates the impact of Ramadan on the returns and the volatility of the stock market indices in India. Design/methodology/approach Initially, the study applies the Ordinary Least Square method to test the day-of-the-week and the month-of-the-year effect of the common and Shariah indices. Next, the study employs the risk-adjusted measurement to examine the underperformance and over-performance of the indices for both the periods. Finally, the study estimates the GARCH (1,1) and GJR-GARCH (1,1) models to observe the impact of Ramadan on the returns and the volatility of the Shariah indices in India. Findings The study finds that an average return of the indices during the Ramadan days are higher than non-Ramadan days. Further, the average returns of the Shariah indices are significantly higher on Wednesday than other days of the week. In addition, the highest and significant mean returns and mean risk-adjusted returns of the indices during the Ramadan days are observed. Finally, the study finds an evidence of the Ramadan effect on the returns and volatility of the indices in India. Originality/value The study observes evidence that the Ramadan effect influences the Shariah indices, but not the common indices in the stock market of the non-Muslim countries. It indicates that the Ramadan creates the positive mood and emotions in the investors buying and selling activities. The study suggests that investors can buy the shares before Ramadan period and sell them during the Ramadan days to get an abnormal return in the emerging markets.

scholarly journals What does and does not correlate with COVID-19 death rates

2020 ◽  
Author(s):  
Christopher R. Knittel ◽  
Bora Ozaltun
Keyword(s):  
Linear Regression ◽  
Public Transportation ◽  
Death Rate ◽  
Negative Binomial ◽  
Statistical Significance ◽  
County Level ◽  
Linear Regression Models ◽  
Death Rates ◽  
Socio Economic Variables ◽  
Home Values

AbstractWe correlate county-level COVID-19 death rates with key variables using both linear regression and negative binomial mixed models, although we focus on linear regression models. We include four sets of variables: socio-economic variables, county-level health variables, modes of commuting, and climate and pollution patterns. Our analysis studies daily death rates from April 4, 2020 to May 27, 2020. We estimate correlation patterns both across states, as well as within states. For both models, we find higher shares of African American residents in the county are correlated with higher death rates. However, when we restrict ourselves to correlation patterns within a given state, the statistical significance of the correlation of death rates with the share of African Americans, while remaining positive, wanes. We find similar results for the share of elderly in the county. We find that higher amounts of commuting via public transportation, relative to telecommuting, is correlated with higher death rates. The correlation between driving into work, relative to telecommuting, and death rates is also positive across both models, but statistically significant only when we look across states and counties. We also find that a higher share of people not working, and thus not commuting either because they are elderly, children or unemployed, is correlated with higher death rates. Counties with higher home values, higher summer temperatures, and lower winter temperatures have higher death rates. Contrary to past work, we do not find a correlation between pollution and death rates. Also importantly, we do not find that death rates are correlated with obesity rates, ICU beds per capita, or poverty rates. Finally, our model that looks within states yields estimates of how a given state’s death rate compares to other states after controlling for the variables included in our model; this may be interpreted as a measure of how states are doing relative to others. We find that death rates in the Northeast are substantially higher compared to other states, even when we control for the four sets of variables above. Death rates are also statistically significantly higher in Michigan, Louisiana, Iowa, Indiana, and Colorado. California’s death rate is the lowest across all states.It is important to understand that this research, and other observational analyses like it, only identify correlations: these relationships are not necessarily causal. However, these correlations may help policy makers identify variables that may potentially be causally related to COVID-19 death rates and adopt appropriate policies after understanding the causal relationship.

scholarly journals THE EFFECT OF RAMADAN MONTH ON STOCK RETURN AND VOLATILITY OF A SHARIA-BASED INDEX

2021 ◽  
Vol 2 (2) ◽  
pp. 119-133
Author(s):  
Arief Rahmatullah ◽  
Putu Anom Mahadwartha ◽  
Endang Ernawati
Keyword(s):  
Linear Regression ◽  
Stock Returns ◽  
Stock Return ◽  
T Test ◽  
Test Results ◽  
Significant Positive Effect ◽  
Dummy Variable ◽  
Average Return ◽  
Study Results ◽  
Positive Effect

This study aims to examine the effect of a religious-related calendar anomaly, namely Ramadan, on stock return and volatility of a Sharia-based index in Indonesia. This study used the GARCH (p,q) method and linear regression to examine the effect of Ramadan on stock returns and volatility, with Ramadan as a dummy variable. This study results show that Ramadan month has a significant positive effect on stock returns, or it can be said that an anomaly occurs during Ramadan month. Meanwhile, volatility during Ramadan month is negative and not significant. This study also exercised a T-test to support the GARCH regression (p,q) and linear regression results. The t-test results show that the average return during Ramadan is higher than in other months. Meanwhile, the average volatility during Ramadan is lower than in other months.

A study about the linearity of trends between extreme and mean temperatures in Argentina

2021 ◽  
Author(s):  
Solange Suli ◽  
Matilde Rusticucci ◽  
Soledad Collazo
Keyword(s):  
Linear Regression ◽  
Linear Relationship ◽  
Statistical Significance ◽  
Coefficient Of Determination ◽  
Maximum Temperature ◽  
Mean Temperature ◽  
Change In The Mean ◽  
The Mean ◽  
Student’S T ◽  
Cold Extremes

<p>Small variations in the mean state of the atmosphere can cause large changes in the frequency of extreme events. In order to deepen and extend previous results in time, in this work we analyzed the linear relationship between extreme and mean temperature (Τ) on a climate change scale in Argentina. Two monthly extreme indices, cold nights (TN10) and warm days (TX90), were calculated based on the quality-controlled daily minimum and maximum temperature data provided by the Argentine National Meteorological Service from 58 conventional weather stations located over Argentina in the 1977–2017 period. Subsequently, we evaluated the relationship between the linear trends of extremes and mean temperature on a seasonal basis (JFM, AMJ, JAS, and OND). Student's T-test was performed to analyze their statistical significance at 5%. Firstly, positive (negative) and significant linear regressions were found between TX90 (TN10) trends and mean temperature trends for the four studied seasons. Therefore, an increase in the Τ-trend maintains a linear relationship with significant increase (decrease) of warm days (cold nights). Moreover, we found that JFM was the one with the highest coefficient of determination (0.602 for hot extremes and 0.511 for cold extremes), implying that 60.2% (51.1%) of the TX90 (TN10) trend could be explained as a function of the Τ-trend by a linear regression. In addition, in the JFM (OND) quarter, the TX90 index increased by 7.02 (6.02) % of days each with a 1 ºC increase in the mean temperature. Likewise, the TN10 index decreased by 4.94 (and 4.99) % of days from a 1ºC increase in the mean temperature for the JFM (AMJ) quarter. Finally, it is worthwhile to highlight the uneven behavior between hot and cold extremes and the mean temperature. Specifically, it was observed that the slopes of the linear regression calculated for the TX90 index and Τ presented a higher absolute value than those registered for the TN10 index and Τ. Therefore, a change in the mean temperature affects hot extremes to a greater extent than cold ones in Argentina.</p>

Part IV Measurement, 16 Legal Issues with Statistical Evidence

2020 ◽  
Author(s):  
Veljanovski Cento
Keyword(s):  
Statistical Analysis ◽  
European Commission ◽  
Best Practice ◽  
Statistical Significance ◽  
Statistical Evidence ◽  
Legal Issues ◽  
Economic Evidence ◽  
Standard Of Proof ◽  
Significance Levels ◽  
General Concern

This chapter examines some of the legal and evidential issues surrounding statistical evidence. Courts are wary of statistical analysis and treat it as complex and difficult to reconcile with legal methods of determining damages based on documentary evidence. Indeed, the problem of estimated averages can conflict with the court’s approach. Moreover, there are statistical, economic, and legal issues surrounding statistical significance. The general concern is that reliance on conventional statistical significance levels may not reflect the legal standard of proof. The court can be assisted by the Practical Guide and Pass-on Guidelines of the European Commission, but also the best practice guidance of competition authorities for the submission of economic evidence.

scholarly journals P003: Door-to-antibiotics and mortality for emergency department patients presenting with septic shock

CJEM ◽  
2019 ◽  
Vol 21 (S1) ◽  
pp. S64
Author(s):  
A. Aguanno ◽  
K. Van Aarsen ◽  
S. Pearce ◽  
T. Nguyen
Keyword(s):  
Emergency Department ◽  
Septic Shock ◽  
Linear Regression ◽  
Hospital Discharge ◽  
Statistical Significance ◽  
Serum Lactate ◽  
Secondary Outcome ◽  
Clinical Equipoise ◽  
Time To Antibiotics ◽  
The Impact

Introduction: We examined our local sepsis patient population, and specifically our most vulnerable patients - those presenting to the emergency department (ED) in septic shock - for variables predictive of survival to hospital discharge. We applied the familiar ED paradigm of, “Door to,” to calculate the impact of time to antibiotics against patient survival to hospital discharge. Methods: Retrospective chart review of patients aged > = 18 years, presenting to tertiary care ED between 01 Nov 2014 and 31 Oct 2015. Patients determined to have sepsis if A) > = 2 SIRS criteria and ED suspicion of infection (ED acquisition of blood/urine cultures or antibiotic administration) and/or B) received ED or Hospital discharge diagnosis of sepsis (ICD-10 diagnostic codes A4xx and R65). Patients sub-classified with septic shock if A) triage SBP < = 90mmHg, B) triage MAP < = 65mmHg or C) serum lactate > = 4mmol/L. “Door Time” was defined as the earliest time recorded for the patient encounter, either the time the patient registered in the Emergency Department, or the triage time. A generalized linear model was performed with a binomial distribution using survival to discharge as the response variable. Age, sex, ED arrival method, time to antibiotics, ED serum lactate and ED serum glucose level were the predictor variables. Results: 13506 patient encounters met inclusion criteria (10980 unique patients). Linear regression of time to antibiotics against survival to hospital discharge failed to achieve statistical significance. Linear regression of the secondary outcome variables achieved statistical significance for age and serum lactate level. Per the model, as age increased by 1 year, the odds of dying prior to hospital discharge increased by 3.8% and as serum lactate increased by 1 mmol/L, odds of dying prior to hospital discharge increased by 11.1%. Conclusion: We found no association between time to antibiotic treatment and mortality. Causal relationships require randomized controlled trials, and this analysis contributes to clinical equipoise.

scholarly journals Analisis January Effect pada Saham Perusahaan LQ-45 di Bursa Efek Indonesia

2013 ◽  
Vol 2 (2) ◽  
pp. 65
Author(s):  
Dwi Cahyaningdyah ◽  
Dhany Kurniawan Putra
Keyword(s):  
Test Data ◽  
Stock Exchange ◽  
T Test ◽  
January Effect ◽  
Purposive Sampling ◽  
Average Return ◽  
Kolmogorov Smirnov

<span class="fontstyle0">The purpose of this research is to examine about the existence of a January Effect in the Indonesia stock exchange. Research samples used purposive sampling. Sample consists 30 companies based on sampling criteria. Analysis of data used one sample kolmogorov-smirnov to test data normality and to test hypotheses using one sample t-test. The result shows January Effect didn’t exist in Indonesia stock exchange in the period 2011- 2012. This can be seen that although the value of average return in January showed significant but the average return was not the highest, the highest average return was occurred on June.</span>

scholarly journals Statistical loadings and latent significance simplify and improve interpretation of multivariate projection models

2018 ◽  
Author(s):  
Pär Jonsson ◽  
Benny Björkblom ◽  
Elin Chorell ◽  
Tommy Olsson ◽  
Henrik Antti
Keyword(s):  
Linear Regression ◽  
Latent Variables ◽  
Dietary Intervention ◽  
Statistical Significance ◽  
Linear Regression Analysis ◽  
True Positive Rate ◽  
Projection Methods ◽  
Systematic Variation ◽  
Model Interpretation ◽  
Metabolomics Study

AbstractMultivariate projection methods are unique in being both multivariable by combining many variables into stronger predictive features (latent variables), and multivariate for being able to model systematic variation both related and orthogonal to an observed response. Orthogonal partial least squares (OPLS) is a versatile multivariate projection method for analysis of correlation, discrimination and effect changes. However, currently OPLS is not fully using its multivariate potential since orthogonal systematic variation is not considered in model interpretation, resulting in univariate interpretation of variable significance. We present a strategy for improved interpretation of OPLS models based upon a post-hoc linear regression analysis that can be used with or without the orthogonal OPLS score(s) as a covariate to make the interpretation multivariate or univariate respectively. By selecting the observed response y or estimated response yhat as a one of the factors in the linear regression the results are related to either of the OPLS loadings w or p. Furthermore, converting the OPLS loading values to statistical t-values creates a direct link to statistical significance. Finally, by applying three different Boolean loadings W, P and W∧P variable significance can be summarized based on three criteria. W and P reveal if the values in w or p respectively are outside the statistical limits with W∧P being the logical conjunction of W and P (significant if outside limits in both W and P). Two examples are used to verify the proposed strategy. First, a synthetic example, simulating a mix of mass spectra, and second a clinical metabolomics study of a dietary intervention. In the simulated example we show that multivariate interpretation gives higher accuracy for estimation of true differences, mainly due to higher true positive rate. Furthermore, we highlight how application of W∧P for summarizing variable significance leads to higher accuracy. For the metabolomics example, we show that a more detailed interpretation, i.e. larger number of significant metabolites of relevance, is obtained using the multivariate interpretation. In summary, the suggested strategy provides means for facilitated interpretation of OPLS models, beyond univariate statistics, and offers a multivariate tool for discovery of biomarker patterns, i.e. latent biomarkers.

scholarly journals Languages, evolution and statistics: human sound systems were shaped by changes in bite configuration. Response to Tarasov & Uyeda (2020)

2020 ◽  
Author(s):  
Damián E. Blasi ◽  
Steven Moran ◽  
Scott R. Moisik ◽  
Paul Widmer ◽  
Dan Dediu ◽  
...  
Keyword(s):  
Linear Regression ◽  
Model Comparison ◽  
Phylogenetic Analyses ◽  
Statistical Evidence ◽  
Causal Models ◽  
Statistical Analyses ◽  
List Type ◽  
Area Variation

AbstractIn Blasi et al. (2019) we have shown, through a series of statistical analyses and models, that human sound systems have been affected by a transition in bite configuration starting from the Neolithic. Tarasov and Uyeda (2020) (henceforth T&U) raise a number of observations in relation to our article. We appreciate T&U’s engagement with our work and their sharing of the code and data of the analyses reported. In brief, their technical comment involves five analyses:Binomial Causal Models (BCM)Linear Regression of across-area variation in labiodentals and subsistencePredictive Posterior Simulations (PPS)Poisson Linear Regression (PLR): model comparisonPhylogenetic AnalysesIn what follows, we show that the discrepancies they report between our findings and theirs are due mostly to ill-specified models, weak (or missing) statistical evidence, and a misinterpretation of our results. After these issues are addressed, we conclude that T&U’s claims do not hold.

Sign in / Sign up

Export Citation Format

Share Document