To be a Quant

Aug 13, 2023

Preparing to be a quant is an exciting journey. There diverse paths to achieve it, since it is characterized for having a set of skills from different branches of knowledge. For sure you’re going to be involved in different topics as: scientific method, finance, behavioral economics, statistics, machine learning, mathematics, probability and technology. Personally the last one was mine.

Managing all this knowledge is no an easy task, mainly if you’re not familiarized with mathematics. Some people asked me what the best starting point is for acomplish the target. My answer is usually the same, it dependes (remember I’m an economist as well :#). There are different types of quants: front office quants, model validation quants, quant researchers, quantitative developers, statistical arbitrage quants, quantitative portfolio manager, and quantitative risk analysts. I’m almost sure will be others in a few years time.

I’ve been working as a front office quant, quantitative risk analyst (market risk) and recently as a model validation quant. In my spare time as a quantitative trader/quantitative portfolio manager of my own portfolio (soon I’ll share my personal track record in another post). In all of them there is logical capabilities and pleasure for maths is a common factor. The ability of creating your own tools a plus, and the creativity to think out of the box will be your personal factor.

• For instance, as a quantitative risk analyst you need to keep an eye on what is happening within portfolios to mitigate the risk, in a daily basis. It may include
  • Assessing your models ensuring they are appropriate for market’s behaviour
  • Calculating measures of risks (i.e. VaR, CVaR/ES, asset haircuts, and so on)
  • Input data analysis and confirmation
  • Asset model development (market data, tools, models) Notice that you must be familiarized with fundamental models and its assumptions. Assumptions are critial. Additionally, you will be working on building tools that other colleagues in other areas use for portfolio risk oversight. Pricing models also bring an opportunity for extra analysis, since those are usually models intended to determine asset’s value in portfolios you are in charge to oversight. A drawback is that the role may turns monotonous when the company is not able to automate recurrent processes due to the business may grow because of diversification which requires expand the analysis to other payoffs.

• As a front quantitative analyst you help traders with different tasks. It encompasses helping to assess strategies by testing with market data and legal constraints (i.e. portfolio’s mandate), until helping them to create mechanisms to integrate front office figures with risk management. In between there are many exciting stuff to do. Just imagine that you are in charge of building models for hedging portfolios. That implies that algorithms for execution must be well known along with asset classes and payoffs thereof. The most of the time the input are time series, reason why knowing time series analysis play a relevant role. Even if portfolio manager is in charge of approve at first instance your models as FO-quant, the responsability also is in your shoulders as designer and implementer. Nobody wants losses in any portfolio using his algorithms. You’ll be getting into in trading vocabulary like order types, trading venues, algorithmic trading straegies, OMS (Order Management Systems), communication protocols (FIX, OUCH, SAIL, iLINK, and others), portfolio mandate, trading book, and so on. A drawback is that when the internal culture with traders is not mature the traders may consider this role as the role for doing the boring stuff and automation, therefore the role may turn into a engineering role … and that is not the goal.

• As a quantitative trader you will produce the halo effect, until you publicly screw it up first time. As some years ago to be a trader is synonymus of highly paid professional. Nowadays it is still true, although it is turning more technical as well given technology develpment in the field. You’ll be involved in building profitable strategies that can be implemented by FO-quants, many times reading and understanding the market data in real time. Thereby, I’d say the most important skill here is to capture consistent patterns through time, and that’s the challenge, or the art. Statistical models might be in tour toolkit, to correctly assess relationships in financial/economic data, and why not alternative data. Besides, you need to be aware about diverse set of strategies as well as for having the capability to identify upfront the ones do not work. The strategies involve where to set limit orders, both TP (Take Profit) and SL (Stop Loss) levels, when to rely more in fundamental than technical analysis, and in the other way around. Nonetheless, technical analysis is likely the most used approach in this field. The drawback is that you may be frustrated if you don’t find a successful strategy which took time to develop. bit lets face it, this guys have a ;level of resilience that many people wish.

• Last but not least, as a quantitative model validator your task as one is to assess the soundness of the models used for different areas. The most common one is quantitative model validator for risk. It implies to assess the soundness on risk models then used for either regulatory requirements or risk management. There are a huge amount of risk metrics for each type of risk (market, credit, liquidiy, counterparty, operational, and so on). Nevertheless, the common factor among them is that those are intensive in both quantitative and statistical methods. Personally this feature is what makes this role quite attractive. You’re not only assessing the models but also learning how developers and first line of defense think to built them. Some times it is plagued of regulation in the banking industry, and honestly for a quant that is not necessarily appealing.

All of those positions have had in common the use of mathematics, statistics, code development, and quantitative finance. At the same time each one has specialized knowledge. So let me list some relevant topics for became a quant.

• Set theory and probability
  • Conditional probability
  • Common probability distributions (Normal, t, Chi, Log-Normal,..)
• Fundamentals of statistics
  • Mean, variance, statistical momentums
  • Hypothesis tests
  • Copulas
• Introduction to algorithms
• Pricing theory
  • Bonds
  • Futures / Forwards
  • Options
  • Swaps
• Risk Management
  • Types of Risk (Market, Credit, Liquidity, so on.)
  • Value at Risk (VaR) : Historical simulation, Parametric (Mean-Variance), Monte Carlo Simulation
  • Expected Shorfall / Conditional Value at Risk (ES / CVaR)
  • Sensitivities
  • Loss Given Default (LGD)
  • Probability of Default (PD)
  • Exposure at Default (EAD)
  • Basel models (I, II, III), and IV soon.
  • Counterparty Clearing House
• Portfolio theory
  • Modern Portfolio Theory (MPT)
  • Capital Asset Pricing Model (CAPM)
  • Arbitrage Pricing Theory (APT)
  • Black-Litterman model
• Quantitative Finance
  • How to build interest rate curves (Nelson-Siegel, Bootstraping, ..)
  • Numerical Methods
  • Stochastic calculus
  • Arbitrage Definition
  • Portfolio Hedging
• Fundamentals on machine learning
• Fundamentals of trading
  • Type of Orders (Limit, Stop-Limit, Market, …)
  • What is trading venue
  • Dark Pools
  • OTC vs Exchange trading

The list is more extensive than this one. I think this one is a good starting point. ENJOY!!