FrmL2_Aspirant
Member
This post was moved from https://forum.bionicturtle.com/thre...iew-of-garps-2017-part-2-practice-exam.10510/
Hi David, my question is not related to this topic, but since you have quoted two different authors above, I am considering you can help me . I want to go through different market risk models that are practically used in the industry, please suggest a good book/site that can be useful in this regard.. thanks!
Question 4 is thematic to the FRM: what are key differences between parametric and non-parametric VaR approaches. In practice, parametric often refers to linear normal VaR models, while non-parametric often refers to historical simulation. (Here a question to tease your understanding of these terms: Is Monte Carlo simulation parametric or non parametric?). But importantly, each is a very broad class with dozens of variations. The FRM continues to lean heavily on Dowd’s summary advantages and disadvantages (copied below; I emphasize those I believe are worth paying careful attention to). In this Question 4, the correct answer depends on SKEWNESS favoring historical simulation. Experienced candidates will note that positive/negative skew can be modeled by non-normal parametric distributions, obviously; so I’m not sure this is a super-strong question. However, it’s obviously the best choice among those given. Choice (A)—”Scarcity of high magnitude loss event”—is EXACTLY why we go to parametric extreme value theory (EVT).
Carol Alexander (MRA Vol IV) introducing Historical Simulation (emphasis mine):
Dowd's EOC summary that informs GARP's Question 4 (emphasis mine)
Hi David, my question is not related to this topic, but since you have quoted two different authors above, I am considering you can help me . I want to go through different market risk models that are practically used in the industry, please suggest a good book/site that can be useful in this regard.. thanks!
Last edited by a moderator: