دانلود مقاله پیش بینی نوسانات بازار سهام پاکستان
خلاصه
1. مقدمه
2. بررسی ادبیات
3. روش شناسی
4. داده ها
5. در نتایج تخمین نمونه
6. خارج از تخمین نمونه
7. نتیجه گیری
اعلامیه منافع رقابتی
قدردانی
منابع
Abstract
1. Introduction
2. Literature review
3. Methodology
4. Data
5. In sample estimation results
6. Out of sample estimation
7. Conclusion
Declaration of competing interest
Acknowledgments
References
چکیده
این مطالعه تاثیر شاخص عدم قطعیت سیاست اقتصادی (EPU) و متغیرهای کلان اقتصادی را بر نوسانات بازار سهام پاکستان با استفاده از مدل GARCH-MIDAS (نمونه‌گیری داده‌های مختلط) بررسی می‌کند. این مدل به ما امکان می دهد مشاهده کنیم که آیا آن متغیرها حاوی اطلاعات ارزشمندی برای پیش بینی نوسانات بازار سهام هستند یا خیر. یافته های تجربی ما چندین نتیجه را نشان می دهد. اول، نتایج خارج از نمونه ما نشان می دهد که شاخص عدم قطعیت سیاست اقتصادی قدرت پیش بینی برای پیش بینی نوسانات بازار سهام پاکستان دارد. دوم، در میان همه متغیرها، قیمت نفت قوی‌ترین پیش‌بینی‌کننده نوسان با مقدار مربع R خارج از نمونه است. سوم، همه متغیرهای کلان اقتصادی از جمله نرخ ارز، نرخ بهره کوتاه مدت، عرضه پول M2، سرمایه گذاری مستقیم خارجی، قیمت طلا، حواله های داخلی، تولید صنعتی و شاخص قیمت مصرف کننده (نمایشی برای تورم) حاوی اطلاعات مفیدی برای پیش بینی نوسانات بازار سهام هستند. با این حال، نرخ بهره بلندمدت یک شاخص بی‌اثر از نوسانات در طول مطالعه دوره نمونه است. در نهایت، متوجه شدیم که اطلاعات پیش‌بینی ترکیبی برای پیش‌بینی نوسانات نیز مفید است.
توجه! این متن ترجمه ماشینی بوده و توسط مترجمین ای ترجمه، ترجمه نشده است.
Abstract
This study examines the impact of the economic policy uncertainty index (EPU) and macroeconomic variables on the volatility of the Pakistan stock market using the GARCH-MIDAS (mixed data sampling) model. The model allows us to observe whether those variables contain valuable information to forecast stock market volatility. Our empirical findings show several outcomes. First, our out-of-sample results show economic policy uncertainty index has predictive power to forecast Pakistan stock market volatility. Second, among all variables, oil prices are the most powerful predictor of volatility with a higher out of sample R square value. Third, all macroeconomic variables including exchange rate, short-term interest rate, money supply M2, foreign direct investment, gold prices, inward remittances, industrial production, and consumer price index (proxy for inflation) contain useful information for stock market volatility forecasting. However, the long-run interest rate is an ineffective indicator of volatility during the sample period study. Finally, we find that the combination forecast information is also useful for volatility forecasting.
Introduction
Volatility in financial markets is an issue of great concern. As a risk measure, volatility considers an important state variable for asset pricing and investment. Additionally, the volatility dynamics are associated with portfolio allocation, risk management, hedging, and options pricing (Christoffersen & Diebold, 2000; Chun et al., 2020; Mei et al., 2017; Morales and Callaghan, 2011; Naeem et al., 2019; Wang and Nishiyama, 2015; Zhang et al., 2020). The higher volatility creates widespread panic and results in disorderly market situations that reduce investors’ confidence and that lead to declines in business investments and economic growth. Therefore, understanding the mechanism of economic dynamics and accurately estimating future volatility is essential.
However, it is prudently observed by financial analysts, policymakers, and market practitioners. Nonetheless, before estimation, it is hard to find the main economic drivers of volatility. The seminal work of SCHWERT and WILLIAM (1989), who posits a time-varying link between stock market uncertainty and macroeconomic variables. Moreover, Paye (2012) examines the strong linear connection between macroeconomic indicators and stock market volatility. Both theoretical and empirical studies confirm that institutional and macroeconomic determinants are potential drivers of financial markets volatility (Engle et al., 2013; Bahloul et al., 2017; Fang et al., 2018; Hsu et al., 2019). The literature shows that financial market volatility and macroeconomic indicators are intrinsically associated (Chen et al., 1986; Mele, 2007; Christiansen et al., 2012).
Conclusion
This research examines the forecasting ability of macroeconomic variables and economic policy uncertainty index (EPU) for Pakistan stock market volatility. The macroeconomic variables include exchange rate, long-run interest rate, money supply (M2), short-term interest rate, remittances, consumer prices index (proxy for inflation), foreign direct investment, industrial production, global oil prices, and gold prices. We use the GARCH-MIDAS model for forecasting short-and long-run volatility. In our empirical analysis, we find some notable results. First, we find economic policy uncertainty index information is useful for Pakistan stock market volatility prediction. Second, we find global indicators oil prices are more powerful predictor of Pakistan market volatility. Moreover, all macroeconomic variables are effective indicators of Pakistan stock market volatility except long run interest rate. In addition, the combination information for all macroeconomic variables is also useful for forecasting volatility. These findings are in line with the literature (Asgharian et al., 2013; Liu & Pan, 2020). However, the abovementioned studies are not directly involved in forecasting Pakistan’s stock market volatility. In addition, we did not find consistent findings in a longer period, such as two months and three months. Overall, results indicate that macroeconomic variables are significant predictors of stock market volatility in Pakistan. Therefore, the government and policymakers should consider these factors in policymaking and stock market volatility estimation.