➡️ Quantitative Finance from Scratch, part 19: value at risk (VaR).
QuantFinanceToGo
TikTok profili37 B takipçi
31.400 görüntüleme #quantfinance #valueatrisk #riskmanagement
➡️ Quantitative Finance from Scratch, part 19: value at risk (VaR). Value at risk is the maximum loss that a portfolio should not exceed over a chosen period, with a chosen probability called the confidence level, under normal market conditions. A VaR of 2% over one day at 99% means: on 99 days out of 100, the loss of a single day stays below 2%. To find it, you sort many past days from the worst loss to the best gain and draw a line where only one day in a hundred is worse. Banks, fund managers and risk teams use it every day to sum up thousands of positions in one number and to set risk limits. J.P. Morgan made it popular in the 1990s. Our picture uses a simulated market with assumed numbers: 100 years of trading days, about 1% a day, with the fat tails from part 18. 99% one-day VaR from the simulated days: 2.7% Same swing on a bell curve: 2.3%, crossed on 396 days instead of the promised 252 At 99.9%: 5.9% against 3.0%, and the bell line is crossed on 186 days instead of about 25 The dangerous problem: VaR says where the worst day in a hundred begins, and nothing about how bad it gets. Past our line, the average loss was 4.2% and the worst day lost 21.5%. That average beyond the line is called expected shortfall, and bank supervisors also moved toward it for market risk after the financial crisis. This video is for education only, and nothing in it is financial advice. #quantfinance #valueatrisk #riskmanagement #finance #python
Yorumlar
Yorumlar TikTok üzerindeki orijinal gönderiden alınmıştır
Henüz yorum yok
