Search Results (142)

The accurate estimation of urban property values is a key challenge for appraisers, market participants, financial institutions, and urban planners. In recent years, machine learning (ML) techniques have emerged as promising tools for price forecasting due to their ability to model complex relationships among variables. However, their application raises two main critical issues: (i) the risk of overfitting, especially with small datasets or with noisy data; (ii) the interpretive issues associated with the “black box” nature of many models. Within this framework, this paper proposes a methodological approach that addresses both these issues, comparing the predictive performance of three ML algorithms—k-Nearest Neighbors (kNN), Random Forest (RF), and the Artificial Neural Network (ANN)—applied to the housing market in the city of Salerno, Italy. For each model, overfitting is preliminarily assessed to ensure predictive robustness. Subsequently, the results are interpreted using explainability techniques, such as SHapley Additive exPlanations (SHAPs) and Permutation Feature Importance (PFI). This analysis reveals that the Random Forest offers the best balance between predictive accuracy and transparency, with features such as area and proximity to the train station identified as the main drivers of property prices. kNN and the ANN are viable alternatives that are particularly robust in terms of generalization. The results demonstrate how the defined methodological framework successfully balances predictive effectiveness and interpretability, supporting the informed and transparent use of ML in real estate valuation. Full article

This study empirically delves into whether residential heating methods significantly affect apartment prices in Uiwang City, a suburban city near the Seoul Metropolitan area, South Korea. Using data from 1256 apartment sales, where both district heating systems (DHSs) and individual heating systems (IHSs) coexist, a hedonic price equation was estimated to analyze the impact of the heating method choices on housing values. Various housing attributes, including physical, locational, and environmental factors, were controlled, and multiple regression models were compared to identify the best-performing specification. The results show that apartments equipped with a DHS are priced, on average, KRW 92 million (USD 72 thousand) higher than those with an IHS. The price difference corresponds to KRW 849 thousand (USD 665) per m² and possesses the statistical significance at the 5% level. Moreover, it is quite meaningful, representing roughly 11.2% of the price of an average apartment. These findings suggest that the use of DHS has a positive effect on apartment prices that reflect consumers’ preferences, beyond its advantages in stable heat supply and energy cost savings. This article provides empirical evidence that DHS can serve as an important urban infrastructure contributing to asset value enhancement. Although this study is based on a specific geographic area and caution must be exercised in generalizing its findings, it reports the interesting finding that residential heating method significantly affects housing prices. Full article

Background/Objectives: CKD-associated pruritus (CKD-aP) is a serious systemic comorbidity occurring in patients with CKD. Despite the burden of CKD-aP, there are limited efficacious treatments available for its management; difelikefalin is the only approved treatment based on its efficacy and safety demonstrated in two clinical studies, namely KALM-1 and KALM-2. This study aimed to evaluate the cost-effectiveness of difelikefalin plus best supportive care (BSC) versus BSC alone when treating moderate-to-severe CKD-aP in patients receiving in-centre haemodialysis, from the perspective of the UK healthcare system. Methods: A de novo lifetime Markov health economic model was built to assess the cost-effectiveness of difelikefalin. The modelled efficacy of difelikefalin was based on data from KALM-1 and KALM-2 pooled at the patient level. The main efficacy driver was the total 5-D Itch scale score. Per-cycle probabilities of changing health states defined by CKD-aP severity were used to derive transition matrices; the model also estimated time-dependent annual probabilities of death and transplant for people on haemodialysis. An increased risk of mortality for modelled patients with very severe, severe, or moderate CKD-aP was applied. Health state utilities and management costs were based on published evidence. Results: Modelled patients treated with difelikefalin were estimated to have a reduced severity of CKD-aP. Consequently, difelikefalin plus BSC was associated with an increased life expectancy of 0.11 years per person and improved HRQoL compared with BSC alone. This translated to higher quality-adjusted life years, at 0.26 per person gained compared to BSC alone. Improved patient outcomes were achieved at an incremental cost of £7814 per person. Conclusions: Overall, at a price of £31.90/vial, difelikefalin was estimated to be a cost-effective treatment for moderate-to-severe CKD-aP at a willingness-to-pay threshold of £30,000/QALY, with conclusions robust to sensitivity analysis. Full article

The performance of machine learning models depends on several factors, including data normalization, which can significantly improve its accuracy. There are many standardization techniques, and none is universally suitable; the choice depends on the characteristics of the problem, the predictive task, and the needs of the model used. This study analyzes how normalization techniques influence the outcomes of real estate price regression models using machine learning to uncover complex relationships between urban and economic factors. Six normalization techniques are employed to assess how they affect the estimation of relationships between property value and factors like social degradation, resident population, per capita income, green spaces, building conditions, and degraded neighborhood presence. The study’s findings underscore the pivotal role of normalization in shaping the perception of variables, accentuating critical thresholds, or distorting anticipated functional relationships. The work is the first application of a methodological approach to define the best technique on the basis of two criteria: statistical reliability and empirical evidence of the functional relationships obtainable with each standardization technique. Notably, the study underscores the potential of machine-learning-based regression to circumvent the limitations of conventional models, thereby yielding more robust and interpretable results. Full article

The usable floor area is one of the key parameters when appraising residential property. In Poland, valuers have to base their analysis on data from the Real Estate Price Register (RCN) in order to value a property. Unfortunately, these data often turn out to be incomplete, especially with regard to floor area, which makes the selection of reference properties difficult and can lead to erroneous valuation results. To address this problem, a study was conducted that used linear models, non-linear models and machine learning algorithms to calculate the floor area of buildings with complex multi-pitched roofs. The analysis was conducted using data sourced from the Database of Topographic Objects (BDOT10k). Three key factors were identified to provide a reliable estimate of usable floor area: the covered area, the height of the building and, optionally, the number of storeys. The results show that the linear model based on the design data achieved an accuracy of 88%, the non-linear model achieved 89% and the machine learning algorithms achieved 93%. For the existing building data from the city of Koszalin, the best model achieved an accuracy of 90%. The estimated values of the usable area of the building designs for the best model on the test set differed on average from the true ones by 8.7 m², while for the existing buildings, the difference was 9.9 m² on average (in both cases, the average relative error was about 7%). Full article

The growing importance of socially responsible investments is causing a rapid increase in the number of various ESG indices. This raises the question of whether the index design matters to stock market investors. The purpose of the article is therefore to analyze the impact of ESG index design on investor decisions motivated by announcements of index reconstitutions. It was assumed that information about company additions to, or deletions from, an index—signaling an improvement in or deterioration of its CSR standards—may be differently interpreted by investors depending on the context provided by the index design. This study used data on the reconstitutions of two ESG indices. One of them, FTSE4Good US, is based on negative screening. Due to its design, membership in it is strongly associated with legitimacy. The other index, DJSI North America, is a best-in-class index which confers a reputation effect. We have applied the event window methodology, which identifies the economic effects of an event by estimating its impact on share prices as reflected in the rate of return. Analysis encompassed 691 events concerning American listed companies in the years 2009–2019, of which 441 were additions and 250 were deletions. It was found that significant investor reactions were triggered only by reconstitutions of the index generating a reputation effect (DJSI). These results indicate that index design does matter. The reactions of investors were positive only when they associated a company’s social commitment with the creation of intangible resources contributing to its competitive advantage. Our results suggest that inclusion in a best-in-class index is more beneficial for a company than in an index based on negative screening. Full article

Heat stress is becoming an increasing concern for dairy farmers due to elevated temperatures and wind shadow caused by rural development. Mechanical ventilation helps mitigate heat stress; however, transitioning from natural to mechanical ventilation increases operational costs. In this study, the number of days with no heat stress, as well as mild, moderate, and severe heat stress, was calculated for Madison, Wisconsin, over the past five years. Monthly milk margins were determined using all milk prices and feed costs from the Dairy Margin Coverage (DMC) program. The goal was to compare the potential reduction in milk margin coverage to the electricity costs of operating ventilation fans. The results indicated that while the five-year average milk margin reduction due to heat stress was USD 20,204 for a 600-head facility, the electricity cost accounted for approximately 42.6% of this amount. However, milk margins fluctuated annually due to volatility in milk and feed markets. For example, in 2021, the reduction in milk margins was estimated at USD 9804, while electricity costs reached USD 8574. It was concluded that in some years, when no severe heat stress occurs, the benefits of ventilation may be close to the expenses. Therefore, adhering to best management practices is critical for minimizing electricity costs while using ventilation fans in dairy operations. Full article

Finite mixture regression identifies homogeneous groups within a sample and computes the regression coefficients in each group. Groups and group coefficients are jointly estimated using an iterative approach. This work extends the finite mixture estimator to the tails of the distribution, by incorporating quantiles and expectiles and relaxing the constraint of constant group probability adopted in previous analysis. The probability of each group depends on the selected location: an observation can be allocated in the best-performing group if we look at low values of the dependent variable, while at higher values it may be assigned to the poorly performing class. We explore two case studies: school data from a PISA math proficiency test and asset returns from the Center for Research in Security Prices. In these real data examples, group classifications change based on the selected location of the dependent variable, and this has an impact on the regression estimates due to the joint computation of class probabilities and class regressions coefficients. A Monte Carlo experiment is conducted to compare the performances of the discussed estimators with results of previous research. Full article

Appropriate operation of the dam reservoir in a hydroelectric power plant (HEPP) is necessary for energy planning, reservoir management, and efficient operation. For good energy planning, the operator needs to make an accurate estimate of the energy production capacity for the next day and plan for production when the energy need is highest. The energy produced in HEPPs depends on the level of water stored in the reservoir, which is directly connected to the reservoir flow. As the water level in the reservoir varies throughout the year depending on climatic conditions, it is important to estimate energy production in order to operate the HEPP most effectively. In this study, the next-day energy production of the HEPP was estimated using a neural network with two hidden layers, each with 10 neurons. A neural network with a hidden layer of 20 neurons was used to estimate future electricity prices and the best hours for market clearing price (MCP). This study found that using short-term training provided the best hourly estimation of MCP, with an average accuracy of 90%; the daily estimation of MCP was ≥95%. Full article

Novel low-cost portable spectrophotometers could be an alternative to traditional spectrophotometers and calibrated RGB cameras by offering lower prices and convenient measurements but retaining high colorimetric accuracy. This study evaluated the colorimetric accuracy of low-cost, portable spectrophotometers on the established color calibration target—RAL Design System Plus (RAL+). Four spectrophotometers with a listed price between USD 100–1200 (Nix Spectro 2, Spectro 1 Pro, ColorReader, and Pico) and a smartphone RGB camera were tested on a representative subset of 183 RAL+ colors. Key performance metrics included the devices’ ability to match and measure RAL+ colors in the CIELAB color space using the color difference CIEDE2000 ΔE. The results showed that Nix Spectro 2 had the best performance, matching 99% of RAL+ colors with an estimated ΔE of 0.5–1.05. Spectro 1 Pro and ColorReader matched approximately 85% of colors with ΔE values between 1.07 and 1.39, while Pico and the Asus 8 smartphone matched 54–77% of colors, with ΔE of around 1.85. Our findings showed that low-cost, portable spectrophotometers offered excellent colorimetric measurements. They mostly outperformed existing RGB camera-based colorimetric systems, making them valuable tools in science and industry. Full article

Serverless computing services are offered by major cloud service providers such as Google Cloud Platform, Amazon Web Services, and Microsoft Azure. The primary purpose of the services is to offer efficiency and scalability in modern software development and IT operations while reducing overall costs and operational complexity. However, prospective customers often question which serverless service will best meet their organizational and business needs. This study analyzed the features, usability, and performance of three serverless cloud computing platforms: Google Cloud’s Cloud Run, Amazon Web Service’s App Runner, and Microsoft Azure’s Container Apps. The analysis was conducted with a containerized mobile application designed to track real-time bus locations for San Antonio public buses on specific routes and provide estimated arrival times for selected bus stops. The study evaluated various system-related features, including service configuration, pricing, and memory and CPU capacity, along with performance metrics such as container latency, distance matrix API response time, and CPU utilization for each service. The results of the analysis revealed that Google’s Cloud Run demonstrated better performance and usability than AWS’s App Runner and Microsoft Azure’s Container Apps. Cloud Run exhibited lower latency and faster response time for distance matrix queries. These findings provide valuable insights for selecting an appropriate serverless cloud service for similar containerized web applications. Full article

One of the key elements in real estate appraisal of residential buildings is the usable area. To determine the monetary value of real estate, appraisers in Poland often rely on transaction data registered in the Real Estate Price Register (REPR). However, the REPR may contain meaningful gaps, particularly on information concerning usable areas. This may lead to difficulties in finding suitable comparative properties, resulting in mispricing of the property. To address this problem, we used linear and nonlinear models to estimate the usable area of buildings with multi-pitched roofs. Utilizing widely available data from the Topographic Objects Database (BDOT10k) based on LiDAR technology, we have shown that three parameters (building’s covered area, building’s height, and optionally the number of storeys) are sufficient for a reliable estimate of the usable area of a building. The best linear model, using design data from architectural offices, achieved a fit of 95%, while the best model based on real data of existing buildings in the city of Koszalin, Poland achieved 92% fit. The best nonlinear model achieved slightly better results than the linear model in the case of design data (better fit by approximately 0.2%). In the case of existing buildings in Koszalin, the best fit was at 93%. The proposed method may help property appraisers determine a more accurate estimation of the usable area of comparative buildings in the absence of this information in the REPR. Full article

Storage profit maximization is based on buying energy at the lowest prices and selling it at the highest prices. The best strategy must thus be based on both accurately predicting the price peak hours and on rightly choosing when to buy and when to sell the stored energy. In this aim, price prediction is crucial, but choosing the prediction model by means of the usual metrics, as the lowest mean squared error, is not an effective solution as the mean squared error computation equally weights the prediction error of all prices, while the focus must be on the higher and lower prices. In this paper, we propose a new metric focused on the correct forecasting of high and low prices so as to allow for a more effective choice among price forecasting models. Results show that the new metric outperforms the standard metrics, allowing for a more accurate estimation of the possible profit for storage (or other trading) activities. Full article

Recently, reservoir computing (RC) has emerged as one of the most effective algorithms to model and forecast volatile and chaotic time series. In this paper, we aim to contribute to the understanding of the uncertainty associated with the predictions made by RC models and to propose a methodology to generate RC prediction intervals. As an illustration, we analyze the error distribution for the RC model when predicting the price time series of several agri-commodities. Results show that the error distributions are best modeled using a Normal Inverse Gaussian (NIG). In fact, NIG outperforms the Gaussian distribution, as the latter tends to overestimate the width of the confidence intervals. Hence, we propose a methodology where, in the first step, the RC generates a forecast for the time series and, in the second step, the confidence intervals are generated by combining the prediction and the fitted NIG distribution of the RC forecasting errors. Thus, by providing confidence intervals rather than single-point estimates, our approach offers a more comprehensive understanding of forecast uncertainty, enabling better risk assessment and more informed decision-making in business planning based on forecasted prices. Full article

This paper uses Long Short-Term Memory (LSTM) networks to predict the stock prices of the Yuanta Taiwan Top 50 ETF (ETF50). To improve the accuracy of the model’s predictions, a calibration procedure called “Short-Term Bias Compensation” (STBC) is proposed to adjust the predicted stock prices. In STBC, the daily prediction error is calculated to estimate the short-term bias (STB) in prediction. Then, the predicted price of its next day will be corrected if this STB has exceeded a certain threshold. In this paper, we apply Genetic Algorithms (GAs) to optimize the parameters used in STBC for providing more confidence in its estimation. Based on these predicted stock prices, we propose a Genetic Fuzzy System (GFS) to determine the trading strategy, with trading points for buying and selling stocks. In GFS, various technical indicators are used to establish the fuzzy rules of the trading strategy, and GAs are used to evolve the best parameters for these fuzzy rules. Our experiments cover over 17 years of data (from 2003 to 2020) for ETF50 to consider black swan events such as the 2020 COVID-19 pandemic, the 2018 US–China trade war, and the 2011 US debt crisis. The first 90% of the data is used as training data, and the last 10% is used as testing data. We use 12 technical indicators of these data as the input of LSTM. The predicted values of LSTM are corrected using STBC and compared to the uncorrected prices. We use Mean Square Error (MSE) to evaluate the prediction accuracy. The results show that STBC can nearly reduce 90% of the prediction error (where MSE drops from 11.5758 to 1.2687). By using GFS with STBC to determine trading points, we achieve a return rate of 32.0%. Full article