Search results

Machine learning methods have recently gained attention in business applications. We will explore the suitability of machine learning methods, particularly support vector regression (SVR) and radial basis function (RBF) approximation, in forecasting company sales. We compare the one-step-ahead forecast accuracy of these machine learning methods with traditional statistical forecasting techniques such as moving average (MA), exponential smoothing, and linear and quadratic trend regression on quarterly sales data of 43 Fortune 500 companies. Moreover, we implement an additive seasonal adjustment procedure on the quarterly sales data of 28 of the Fortune 500 companies whose time series exhibited seasonality, referred to as the seasonal group. Furthermore, we prove a mathematical property of this seasonal adjustment procedure that is useful in interpreting the resulting time series model. Our results show that the Gaussian form of a moving RBF model, with or without seasonal adjustment, is a promising method for forecasting company sales. In particular, the moving RBF-Gaussian model with seasonal adjustment yields generally better mean absolute percentage error (MAPE) values than the other methods on the sales data of 28 companies in the seasonal group. In addition, it is competitive with single exponential smoothing and better than the other methods on the sales data of the other 15 companies in the non-seasonal group.

Purpose: Investors and portfolio managers can earn profitably when they correctly predict when stock prices will go up or down. For this reason, it is crucial to know the effect levels of the factors that affect stock prices. In addition to macroeconomic factors, the psychological behavior of investors also affects stock prices. Therefore, the study aims to reveal the different sensitivity levels of the stock index against macroeconomic and psychological factors.

Design/Methodology/Approach: In this study, dollar rate (USD), euro rate (EURO), time deposit interest rate (IR), gold price (GOLD), industrial production index (IPI), and consumer price index (CPI) (inflation (INF)) were used as macroeconomic factors, while Consumer Confidence Index (CCI) and VIX Fear Index (VIX) were used as psychological factors. In addition, the BIST-100 index, which is listed in Borsa Istanbul, was used as the stock index. The sensitivity of the stock index to macroeconomic and psychological factors was investigated using the Multivariate Adaptive Regression Spline (MARS) method using data from January 2012 to October 2020.

Findings: In the analyses performed using the MARS method, the coefficients of INF, USD, EURO, IR, CCI, and VIX Index were found to be statistically significant and effective on the stock index. Among these variables, INF has the highest effect on stocks. It is followed by USD, IR, EURO, CCI, and VIX. GOLD and IPI variables did not show statistical significance in the model. The most important difference of the MARS model from other regressions is that each factor’s effect on the stock index is analyzed by separating it according to the value of the factor. According to the results obtained from the MARS model: (1) it has been determined that USD, EURO, IR, and CPI have both positive and negative effects on the stock market index and (2) CCI and VIX have been found to have negative effects on stocks. These results provide essential information about how investors who plan to invest in the stock index should take into consideration different macroeconomic and psychological values.

Originality/value: This study contributes to the literature as it is one of the first studies to examine the effects of factors affecting the stock index by decomposing it according to the values it takes. Also, this study provides additional information by listing the factors affecting the stock index in order of importance. These results will help investors, portfolio managers, company executives, and policy-makers understand the stock markets.

Demand seasonality in the U.S. Imported Beer industry is common. The financial cycles of the past decade brought some extreme fluctuations to industry demand, which was trending upward. This research extends previous work in this area by comparing seasonal forecasting models for two time periods: 1999–2007 and 1999–2012. The previous study (Kros & Keller, 2010) examined the 1999–2007 time frame while this study extends their model using the new data. Models are developed within Excel and include a simple yearly model, a semi-annual model, a quarterly model, and a monthly model. The results of the models are compared and a discussion of each model’s efficacy is provided. While, the models did do a good job forecasting U.S. Import Beer sales from 1999 to 2007 the economic downturn starting in 2007 was deleterious to some models continued efficacy. When the data from the downturn is accounted for it is concluded that the seasonal models presented are doing an overall good job of forecasting U.S. Import Beer Sales and assisting managers in shorter time frame forecasting.

This chapter presents an Excel-based regression analysis to forecast seasonal demand for U.S. Imported Beer sales data. The following seasonal regression models are presented and interpreted including a simple yearly model, a quarterly model, a semi-annual model, and a monthly model. The results of the models are compared and a discussion of each model's efficacy is provided. The yearly model does the best at forecasting U.S. Import Beer sales. However, the yearly does not provide a window into shorter-term (i.e., monthly) forecasting periods and subsequent peaks and valleys in demand. Although the monthly seasonal regression model does not explain as much variance in the data as the yearly model it fits the actual data very well. The monthly model is considered a good forecasting model based on the significance of the regression statistics and low mean absolute percentage error. Therefore, it can be concluded that the monthly seasonal model presented is doing an overall good job of forecasting U.S. Import Beer Sales and assisting managers in shorter time frame forecasting.

In the last few decades, there has been growing interest in forecasting with computer intelligence, and both fuzzy time series (FTS) and artificial neural networks (ANNs) have gained particular popularity, among others. Rather than the conventional methods (e.g., econometrics), FTS and ANN are usually thought to be immune to fundamental concepts such as stationarity, theoretical causality, post-sample control, among others. On the other hand, a number of studies significantly indicated that these fundamental controls are required in terms of the theory of forecasting, and even application of such essential procedures substantially improves the forecasting accuracy. The aim of this paper is to fill the existing gap on modeling and forecasting in the FTS and ANN methods and figure out the fundamental concepts in a comprehensive work through merits and common failures in the literature. In addition to these merits, this paper may also be a guideline for eliminating unethical empirical settings in the forecasting studies.

The relationship between electricity demand and weather in the United States has been studied as of late due to increased demand, de-regulation, and new pricing models. The influence of weather or seasonality in energy consumption, particularly electricity demand, has been widely researched. A significant scientific interest in the seasonality of energy consumption has led to an important number of papers exploring the role of weather variability and change on energy consumption. Most of these papers model demand as a function of seasonal climate factors.

The goal of this research is a broad examination of monthly residential electricity demand for a region of the mid-Atlantic using Excel and step-wise regression. This is achieved by using a sequence of models built in Excel in which different patterns are gradually introduced in the estimations. Data over a seven-year period is utilized. A backward elimination step-wise regression analysis is employed to determine which independent variables best model the data. Initial independent variables included high monthly temperature, low monthly temperature, time, year, month, seasonal quarter, and introduction of a “green” tax credit for solar and wind energy.

Models for forecasting the electricity demand and the predictive power of these models is assessed. The work is organized as follows: Data description and the methodology, trend and the seasonality of electricity usage in the mid-Atlantic region, the predictive power and seasonality of the models, and main conclusions drawn from the study.

Product success is contingent on forecasting when a product is needed and how it should be offered. Forecasting accuracy is contingent on the correct forecasting technique. Using supermarket data across two product categories, this chapter shows that using a bevy of forecasting methods improves forecasting accuracy. Accuracy is measured by the mean absolute percentage error. The optimal methods for one consumer goods product may be different than for another. The best model varied from sophisticated, most such as autoregressive integrated moving average (ARIMA) and Holt–Winters to a random walk model. Forecasters must be proficient in multiple statistical techniques since the best technique varies within a categories, variety, and product size.

Purpose – This chapter analyses the various themes connected with cycling's current situation and future prospects which have emerged through the previous 10 chapters, and elaborates the need for a ‘bicycle system’ which is capable of achieving a ‘revolution’ in cycling.

Approach – The chapter draws on previous chapters, as well as the results of recently completed research into the state of cycling across urban England.

Findings – Cycling remains marginalised, but its current rise in status across some of the world's cities offers grounds for optimism about its future contribution to sustainability objectives. The bicycle's rise in status is currently both elitist and, potentially, a passing fashion; the challenge is to make it both more democratic and durable.

Practical implications – In the mould of ‘common endeavours’ outlined in the World Commission Report on Environment and Development ‘Our Common Future’, the authors propose building a ‘bicycle system’ to ensure the bicycle can play a full role in the transition to (especially urban) sustainability and outline possible principles for, pathways towards, and components and characteristics of, a bicycle system.

Social implications – The chapter aims to influence broader debates, and importantly it needs to influence political discourse, about the changes required to assist in the transition to greater urban transport sustainability, and specifically to discourage car use whilst encouraging use of the bicycle for short urban journeys.

Value of paper – The authors provide an analysis of the current constraints on cycling, and a case for simultaneously assembling a ‘bicycle system’ as the means of transitioning urban transport towards sustainability, whilst at the same time disassembling the current system that allows cars to predominate.