Pub Date : 2023-09-03DOI: 10.18187/pjsor.v19i3.4247
Chantha Wongoutong
Herein, a modified weighting for combined forecasting methods is established. These weights are used to adjust the correlation coefficient between the actual and predicted values from five individual forecasting models based on their correlation coefficient values and ranking. Time-series datasets with three patterns (stationary, trend, or both trend and seasonal) were analyzed by using the five individual forecasting models and three combined forecasting methods: simple-average, Bates-Granger, and the proposed approach. The MAPE and RMSE results indicate that the proposed method outperformed the others, especially when the time-series pattern was stationary and improved the forecasting accuracy of the worst and best individual forecasting models by 35–37% and 7–10%, respectively. Moreover, the proposed method showed improvements in MAPE and RMSE of around 18–20% and 9–11% compared to the simple-average and Bates-Granger methods, respectively. In addition, the combined forecasting methods outperformed the individual forecasting models when analyzing non-stationary data. Remarkably, the performances of the proposed and Bates-Granger methods were almost the same, with improvements in MAPE and RMSE in the range of 1–2% on average. Therefore, the proposed method for creating weights based on the correlation coefficients of the individual forecasting models greatly improves combined forecasting methods.
{"title":"A modified weighting system for combined forecasting methods based on the correlation coefficients of the individual forecasting models","authors":"Chantha Wongoutong","doi":"10.18187/pjsor.v19i3.4247","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.4247","url":null,"abstract":"Herein, a modified weighting for combined forecasting methods is established. These weights are used to adjust the correlation coefficient between the actual and predicted values from five individual forecasting models based on their correlation coefficient values and ranking. Time-series datasets with three patterns (stationary, trend, or both trend and seasonal) were analyzed by using the five individual forecasting models and three combined forecasting methods: simple-average, Bates-Granger, and the proposed approach. The MAPE and RMSE results indicate that the proposed method outperformed the others, especially when the time-series pattern was stationary and improved the forecasting accuracy of the worst and best individual forecasting models by 35–37% and 7–10%, respectively. Moreover, the proposed method showed improvements in MAPE and RMSE of around 18–20% and 9–11% compared to the simple-average and Bates-Granger methods, respectively. In addition, the combined forecasting methods outperformed the individual forecasting models when analyzing non-stationary data. Remarkably, the performances of the proposed and Bates-Granger methods were almost the same, with improvements in MAPE and RMSE in the range of 1–2% on average. Therefore, the proposed method for creating weights based on the correlation coefficients of the individual forecasting models greatly improves combined forecasting methods.","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":" ","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47256562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-03DOI: 10.18187/pjsor.v19i3.3983
Ramesh Ganesan, Mathavan N
The objective of this article is studying on cost and time minimization of interval transportation problem (ITP) by using Best Candidate Method (BCM), Improved ASM method (IASM), ASM method, Zero Suffix Method (ZSM) and Zero Point Method (ZPM) with new interval arithmetic operations. We have obtained a better optimum result campared with existing methods available in the literature. The problems considered in this article are solved by the above listed methods without converting them into classical transportation problems. A comparative results are also given.
{"title":"A Class of Methods Using Interval Arithmetic Operations for Solving Multi–Objective Interval Transportation Problems","authors":"Ramesh Ganesan, Mathavan N","doi":"10.18187/pjsor.v19i3.3983","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.3983","url":null,"abstract":"The objective of this article is studying on cost and time minimization of interval transportation problem (ITP) by using Best Candidate Method (BCM), Improved ASM method (IASM), ASM method, Zero Suffix Method (ZSM) and Zero Point Method (ZPM) with new interval arithmetic operations. We have obtained a better optimum result campared with existing methods available in the literature. The problems considered in this article are solved by the above listed methods without converting them into classical transportation problems. A comparative results are also given.","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":"1 1","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"67519602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-03DOI: 10.18187/pjsor.v19i3.4113
A. Adetunji, Shamsul Rijal Muhammad Sabri
In this study, a new three-parameter mixed Poisson Cubic Rank Transmuted New Weighted Exponential Distribution is proposed. The new discrete distribution is obtained by mixing the Poisson distribution with a newly obtained Cubic Rank Transmuted New Weighted Exponential Distribution. Various shapes and mathematical properties of both mixing distribution and the new count distribution are examined. Special cases of the new proposition are also identified. The distribution along with its special cases and other count distributions are assumed for skewed and dispersed count observations. The maximum likelihood estimation is used to provide estimates for the parameters of all examined distributions. Results show that the new proposition along with some of its special cases provide good fit for all the examined data.
{"title":"Mixed Poisson Transmuted New Weighted Exponential Distribution with Applications on Skewed and Dispersed Count Data","authors":"A. Adetunji, Shamsul Rijal Muhammad Sabri","doi":"10.18187/pjsor.v19i3.4113","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.4113","url":null,"abstract":"In this study, a new three-parameter mixed Poisson Cubic Rank Transmuted New Weighted Exponential Distribution is proposed. The new discrete distribution is obtained by mixing the Poisson distribution with a newly obtained Cubic Rank Transmuted New Weighted Exponential Distribution. Various shapes and mathematical properties of both mixing distribution and the new count distribution are examined. Special cases of the new proposition are also identified. The distribution along with its special cases and other count distributions are assumed for skewed and dispersed count observations. The maximum likelihood estimation is used to provide estimates for the parameters of all examined distributions. Results show that the new proposition along with some of its special cases provide good fit for all the examined data.","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":" ","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47473926","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-03DOI: 10.18187/pjsor.v19i3.4230
A. Afify, R. R. Pescim, G. Cordeiro, H. A. Mahran
A new weighted exponentiated-exponential distribution is proposed to model financial data. It has heavy-tailed behavior which is suitable for data with right tails. Some actuarial measures for the new model are determined, and simulations are reported. Its parameters are estimated using nine approaches including a Bayesian method. A new Log-WEx-Exponential regression model is defined for right censored data. The importance of the new models is proved by applications to financial data.
{"title":"A New Heavy-Tailed Exponential Distribution: Inference, Regression Model and Applications","authors":"A. Afify, R. R. Pescim, G. Cordeiro, H. A. Mahran","doi":"10.18187/pjsor.v19i3.4230","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.4230","url":null,"abstract":"A new weighted exponentiated-exponential distribution is proposed to model financial data. It has heavy-tailed behavior which is suitable for data with right tails. Some actuarial measures for the new model are determined, and simulations are reported. Its parameters are estimated using nine approaches including a Bayesian method. A new Log-WEx-Exponential regression model is defined for right censored data. The importance of the new models is proved by applications to financial data.","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":" ","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46460036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-03DOI: 10.18187/pjsor.v19i3.4089
Mohamed K. A. Refaie, Asmaa Ayoob Yaqoob, Mahmoud Ali Selim, Emadeldin I. A. Ali
In this study, the authors of the current work describe a novel exponentiated Weibull distribution that they have invented. The study was written by the writers of the current work. It is required to analyze those properties once the pertinent mathematical properties have been derived. In addition to the dispersion index, the anticipated value, variance, skewness, and kurtosis are also statistically examined. The dispersion index is likewise examined. Other beneficial shapes that the new density can assume include "bathtub," "right skewed," "bimodal and left skewed," "unimodal and left skewed," and "bimodal and right skewed." Additionally, these forms can be merged to create a "bathtub." The term "bathtub (U-HRF)," "constant," "monotonically increasing," "upside down-increasing (reversed U-increasing)," "J-HRF," "upside down-constant," "increasing-constant," or "upside down (reversed U)" may be used to describe the new rate of failure. The greatest likelihood method's efficiency is assessed via graphical analysis. The main measures for this procedure’s evaluation are biases and mean squared errors. The reader is given a scenario that graphically displays the adaptability and value of the innovative distribution through the use of three separate sets of actual data.
{"title":"A Novel Version of the Exponentiated Weibull Distribution: Copulas, Mathematical Properties and Statistical Modeling","authors":"Mohamed K. A. Refaie, Asmaa Ayoob Yaqoob, Mahmoud Ali Selim, Emadeldin I. A. Ali","doi":"10.18187/pjsor.v19i3.4089","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.4089","url":null,"abstract":"In this study, the authors of the current work describe a novel exponentiated Weibull distribution that they have invented. The study was written by the writers of the current work. It is required to analyze those properties once the pertinent mathematical properties have been derived. In addition to the dispersion index, the anticipated value, variance, skewness, and kurtosis are also statistically examined. The dispersion index is likewise examined. Other beneficial shapes that the new density can assume include \"bathtub,\" \"right skewed,\" \"bimodal and left skewed,\" \"unimodal and left skewed,\" and \"bimodal and right skewed.\" Additionally, these forms can be merged to create a \"bathtub.\" The term \"bathtub (U-HRF),\" \"constant,\" \"monotonically increasing,\" \"upside down-increasing (reversed U-increasing),\" \"J-HRF,\" \"upside down-constant,\" \"increasing-constant,\" or \"upside down (reversed U)\" may be used to describe the new rate of failure. The greatest likelihood method's efficiency is assessed via graphical analysis. The main measures for this procedure’s evaluation are biases and mean squared errors. The reader is given a scenario that graphically displays the adaptability and value of the innovative distribution through the use of three separate sets of actual data.","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":" ","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45231633","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-03DOI: 10.18187/pjsor.v19i3.4157
Dari Sani
In this paper, an EPQ model for items that exhibit delay in deterioration is developed. It is assumed that there is no demand and no deterioration during production buildup period. Demand starts immediately after production but no deterioration. Then a period of deterioration sets in until the stock finishes. It is also supposed that the cost of a unit product is inversely related to the rate of demand and directly related to the process reliability (as assumed by Tripathy et al. (2015) and modified by Dari and Sani (2015)). The demand before deterioration sets in is quadratic time dependent while demand after deterioration sets in is a constant. Shortages are allowed and partially backordered. A numerical model is given to compare the simulation model and the statistical analysis conducted on the model to see the effect of measurement changes in other system parameters.
{"title":"An EPQ Model for Delayed Deteriorating Items with Reliability Consideration, Quadratic Demand and Shortages","authors":"Dari Sani","doi":"10.18187/pjsor.v19i3.4157","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.4157","url":null,"abstract":"In this paper, an EPQ model for items that exhibit delay in deterioration is developed. It is assumed that there is no demand and no deterioration during production buildup period. Demand starts immediately after production but no deterioration. Then a period of deterioration sets in until the stock finishes. It is also supposed that the cost of a unit product is inversely related to the rate of demand and directly related to the process reliability (as assumed by Tripathy et al. (2015) and modified by Dari and Sani (2015)). The demand before deterioration sets in is quadratic time dependent while demand after deterioration sets in is a constant. Shortages are allowed and partially backordered. A numerical model is given to compare the simulation model and the statistical analysis conducted on the model to see the effect of measurement changes in other system parameters.","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":" ","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48311375","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-03DOI: 10.18187/pjsor.v19i3.4143
S. Bharali, Jiten Hazarika, Kuldeep Goswami
A mathematical model is a mathematical connection that describes some real-life scenario. To handle real-world problems securely and effectively, simulation modelling is required. In this article, the author investigates the stochastic regression model scenario in which the dependent and independent variables in a linear regression model follow a distribution. We assume that the dependent and independent variables both exhibit Type I Extreme Value Distribution. The estimators are then derived using the Modified Maximum Likelihood (MML) estimation method. In accordance with this, a hypothesis testing technique is developed.
{"title":"Marginal and Conditional both Extreme Value Distributions: A Case of Stochastic Regression Model","authors":"S. Bharali, Jiten Hazarika, Kuldeep Goswami","doi":"10.18187/pjsor.v19i3.4143","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.4143","url":null,"abstract":"A mathematical model is a mathematical connection that describes some real-life scenario. To handle real-world problems securely and effectively, simulation modelling is required. In this article, the author investigates the stochastic regression model scenario in which the dependent and independent variables in a linear regression model follow a distribution. We assume that the dependent and independent variables both exhibit Type I Extreme Value Distribution. The estimators are then derived using the Modified Maximum Likelihood (MML) estimation method. In accordance with this, a hypothesis testing technique is developed.","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":" ","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44511744","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-03DOI: 10.18187/pjsor.v19i3.2877
Preeti Patidar, H. P. Singh
In this paper we have suggested a class of estimators of population mean of sensitive variable under optional randomized response technique as reported in Gupta et al  (2014). We have obtained the mean squared error (MSE) of the suggested class of estimators up to the first order of approximation. The optimum conditions are obtained at which the (MSE) of the proposed class of estimators is minimum. An empirical study is carried out to show the performance of the suggested class of estimators over existing estimators .It is found that the performance of proposed class of estimators is better than the existing estimators including Grover and Kaur (2019).
{"title":"An Improved Class of Estimators Of Population Mean of Sensitive Variable Using Optional Randomized Response Technique","authors":"Preeti Patidar, H. P. Singh","doi":"10.18187/pjsor.v19i3.2877","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.2877","url":null,"abstract":"In this paper we have suggested a class of estimators of population mean of sensitive variable under optional randomized response technique as reported in Gupta et al  (2014). We have obtained the mean squared error (MSE) of the suggested class of estimators up to the first order of approximation. The optimum conditions are obtained at which the (MSE) of the proposed class of estimators is minimum. An empirical study is carried out to show the performance of the suggested class of estimators over existing estimators .It is found that the performance of proposed class of estimators is better than the existing estimators including Grover and Kaur (2019).","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":" ","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42348621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, separate and combined ratio type estimators have been proposed in presence of non-response for estimating the population mean under stratified random sampling when the non-response occurs both on study and the auxiliary variables and the population mean of the auxiliary variable is unknown. The expressions for the biases and mean square errors (MSEs) of the proposed estimators have been derived to the first order of approximation. The proposed estimators have been compared with the other existing estimators using MSE criterion, and the condition under which the proposed estimators perform better than existing estimators have been obtained. In addition to the theoretical research, an empirical study was conducted.
{"title":"Assessing the Effect of Non-response in Stratified Random Sampling using Enhanced Ratio Type Estimators under Double Sampling Strategy.","authors":"Zakir Hussain Wani, S.E.H. Rizvi, 𝑛𝑥̅ 𝑠𝑡∗, 𝛾, 𝑛 𝑛′𝑥̅−, 𝑦̅ 𝑧𝑟𝑐𝑝∗, 𝑦̅, 𝑠𝑡∗, 𝑛 ′ − 𝑛 𝑛𝑥̅ 𝑠𝑡, 𝑥̅ 𝑠𝑡𝑆, 𝑥̅ 𝑠𝑡∗′, 𝑠𝑡𝑆, 𝑋̅, 𝑛 1𝑛′−, 1 𝑋̅, 𝑡, 𝑡 𝜉1𝑠𝑡∗′−, 𝜉, 𝑋̅ − 𝑋̅, − 𝑌̅ 𝑌̅, 𝑡 −, 𝜉 1𝑠𝑡∗, 𝑡 𝜉1𝑠𝑡∗′, 𝑦 𝑧𝑟𝑐𝑝∗","doi":"10.18187/pjsor.v19i3.4063","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.4063","url":null,"abstract":"In this paper, separate and combined ratio type estimators have been proposed in presence of non-response for estimating the population mean under stratified random sampling when the non-response occurs both on study and the auxiliary variables and the population mean of the auxiliary variable is unknown. The expressions for the biases and mean square errors (MSEs) of the proposed estimators have been derived to the first order of approximation. The proposed estimators have been compared with the other existing estimators using MSE criterion, and the condition under which the proposed estimators perform better than existing estimators have been obtained. In addition to the theoretical research, an empirical study was conducted.","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":" ","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44554927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-03DOI: 10.18187/pjsor.v19i3.3885
Dimpal Pathak, P. Hazarika, Subrata Chakraborty, Jondeep Das, G. G. Hamedani
This paper considers a new family of the trimodal skew logistic distributions. Some properties of this distribution, including moments, moments generating function, entropy, maximum likelihood estimates of parameters and some other properties, are presented. A simulation study is conducted to examine the performance of the parameters. Numerical optimization is carried out via two real-life datasets. Results show that the new distribution is better fitted in terms of these datasets among logistic, skew logistic and alpha skew logistic distributions based on the value of AIC and BIC.
{"title":"Modeling Tri-Model Data With a New Skew Logistic Distribution","authors":"Dimpal Pathak, P. Hazarika, Subrata Chakraborty, Jondeep Das, G. G. Hamedani","doi":"10.18187/pjsor.v19i3.3885","DOIUrl":"https://doi.org/10.18187/pjsor.v19i3.3885","url":null,"abstract":"This paper considers a new family of the trimodal skew logistic distributions. Some properties of this distribution, including moments, moments generating function, entropy, maximum likelihood estimates of parameters and some other properties, are presented. A simulation study is conducted to examine the performance of the parameters. Numerical optimization is carried out via two real-life datasets. Results show that the new distribution is better fitted in terms of these datasets among logistic, skew logistic and alpha skew logistic distributions based on the value of AIC and BIC.","PeriodicalId":19973,"journal":{"name":"Pakistan Journal of Statistics and Operation Research","volume":" ","pages":""},"PeriodicalIF":1.5,"publicationDate":"2023-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43378624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：