lilac bushIt's not barbecue, but today I had the pleasure for the first time of tasting lilacs. I was inspired by a kid's book that said that to find out why bees like the taste of sweet things you should squeeze the nectar of some nettle plants. I took Aubrey to the Bronx Botanical Gardens and we smelled the lilacs. By far the best-smelling were the hyacinth lilacs. To me it's the best smell in the world, and it stops time and elicits every romantic spring personage that one could ever imagine.

Inspired by the reverie, I couldn't resist tasting a number of the small flowers. I found that the white ones on the top of the trees were superior in taste and smell to the red ones, especially lilac poincare and lilac common. The taste is like a mixture of raspberries and sweet peas. A slight tartness at first, but then a beautiful saladly green with sweet overtones.

I've seen some recipes for lilacs subsequently, but surprisingly nothing on the actual taste of lilacs, indeed almost a googlewhack. I highly recommend that all speculators take a break from their trading before or after the daily fray and sample a few in their area.

Phil McDonnell writes:

As a home gardener I have been amazed at the number of flowers that can be eaten and are considered delicious gourmet treats. For example, zucchini flowers. Zucchini plants have two types of flowers, male and female. The male flowers stand erect and tall as is only proper. The female flowers are short stemmed and demurely lower. Even though the two flowers look similar, I find it very easy to remember the difference.

The taller male flowers are the first to be found by the bees. They next visit the lower females, thus facilitating fruiting. If bees are lacking, then the higher male flowers can pollinate the lower female flowers simply through wind action. After the male's job is done the gourmet can harvest the male flowers for a real treat. Just sautee in butter, salt, garlic and onions. Each female will give you a zucchini.

Vincent Andres adds:

Reading Vic's post, I was thinking exactly on that: "fleur-de-courgette" and I didn't know it translated as "zucchini flowers". Thanks Phil. I'll be growing some in my potager every year. Beignet-de-fleur-de-courgette is a terribly good (and not so difficult) flower recipe. Here is a video.  I expect to have my own olive oil in the coming years to made it 100% with raw home products. 

Bruno Ombreux writes:

May I recommend a French delicacy: Crystallized violet flowers.

They are very easy to make. These are the flowers you use.





Speak your mind

2 Comments so far

  1. Don Chu on April 19, 2010 11:31 am

    Every kid must had been told not to taste the plants and flowers growing outside; but how many of us could resist the sweet nectar from a beautiful flower, whose single drop of heaven-sent cordial is as precious and delicate as morning dew and manna.

    I remember as a child, adults warning us especially against the supposedly poisionous Flame of the Woods or Ixora plant, with its bunches of wild-fiery small flowers. But we went ahead anyway:

    Pluck a single fully-bloomed flower, and from the bottom of the petals, carefully poke through the base of the petals without severing the stem and puncturing the nectary near the base.
    If carried out properly, the style, a filament-like string should be exposed. Pull on the delicate style very gently, which should pull the attached stigma at the other end along.
    When the stigma gets caught at the narrow opening at the nectary, it should release the nectar, forming a single drop of sweet nectar flowing down the style.
    Lick. Enjoy.

  2. douglas roberts dimick on April 22, 2010 9:10 am

    Sensitivity Analysis and Trading Patterns: How to Stop and Smell the Flowers?

    This past Saturday, upon invitation by SHU's Foreign Affairs Office, I joined 15 Chinese and foreign members of our university for a day at the Expo 2010 Shanghai Botanical Garden (see[L]Skins/waishie1/waishie1 ).

    After lunch and upon waking from a nap under a tree, I sat up and, watching people wander about, it dawned on me that market analysis is like observing people viewing, smelling, touching, even picking flowers at such venues as a botanical garden.

    Like securities, flowers are planted, grow, bloom, die, even some are reborn. They survive and perish with conditions and events that are natural (or elemental, such as wind, rain, etc.) as well as human (or artificial, such as chemical and biologic interaction or exposure). There are also all kinds of sections… just like the different electronic exchange markets.

    The question then becomes, before picking, how may one come to view potential selections, be it flowers or securities?

    Checking my email the day after our trip, I noticed that I started receiving email notifications from my account about the Quant Linked Group (QLG: Wall Street Quants). Thus, as with moving among Shanghai’s tulips and New York’s lilacs, I highlight here comments offered that appear particularly interesting for further consideration.

    “Why is sensitivity analysis dangerous?

    1. The notion of freezing (i.e. controlling) all variables that impact an outcome, except one is not supposed to represent reality. However, it is a standard method of science to understand the effect of a single variable in a multivariate situation. It is most applicable when the effects of all variables are independent…often not the case in financial situations.

    2. In fact, in sensitivity analysis one assumes that for "all things being equal" something happens. Clearly, in life it is very difficult to find situations in which all things are equal, so that some hypothesis may be tested. The danger I see behind similar analytical tools is that we use something that declaredly doesn't reflect reality to make decisions in the real world. How can that not be dangerous or, at least, risky?

    3. I would add another flaw in sensitivity analysis: the real risk on a given variable (also called "risk factor") is with a large shift of the variable. In this case, the sensitivity analysis assumes that the impact is proportional to the size of the shift. For instance, the impact of a 300 bp tightening of credit spreads is 300 x PVO1. There is nothing more wrong than that. Nonlinearities are often very strong. It doesn't mean extrapolations are impossible, but they should be made with care, using all available info and not just extrapolating a very small move.

    4. Well, scenario analysis (a broader conceptual extension of sensitivity analysis) is just a tool. The issue of impracticality (as raised by Jacek and Raphael) is actually due our inability to not exactly understand (and quantify) the inter-relationship of various variables. Scenario analysis is, in fact, very simple in concept but very useful in practice. It allows you to develop scenarios, be it based on historical data, coming out of Early Warning Indicators or simply hypotheticals. There is absolutely no need to assume all things being equal, if you can model the interaction of variables (linear or non-linear relations). Hypothetical scenarios could be converted "near to really extreme probables" by conducting reverse-stress testing concept.

    In fact, reverse stress-testing, meaning reaching towards scenarios backward from unacceptable BS/P&L conditions (usually going beyond the usual risk-appetite) of an organization, helps go back to take note of the risk factors from reality point of view.

    5. Sensitivity analysis as performed is usually an application of what's known as a Taylor expansion in several variables. The weakness of the method is its poor performance for many functions of practical interest, due to either discontinuities, lack of proper smoothness, or highly non-linear behavior that makes higher order partial derivatives have material impact on the function's behavior. Problems with sensitivity analysis show up in many applications, such as fluid dynamics. In Finance, the need for additional "Greek letters" besides delta is an example of the deficiencies of sensitivity analysis, as Raphael points out. The quirky behavior of some complex options is only explained by more partial derivatives than Finance has Greek letter names for. Taylor's formula works well for typical functions found in calculus textbooks, but practitioners need to be more cautious.

    6. Sensitivity Analysis itself is not the problem; it is merely a tool in the arsenal for considering the impact small changes in parameters. However it must be combined with a consideration of extreme scenarios (i.e. what happens if interest rates double, what happens if market drops 20%, what happens if short volatility quadruples, what was the worst case scenario in the past if data available, and what was the most like worst case scenario in a similar instrument if there is no data available). For example scenario analysis would not have helped much with say Lehman brothers but an understanding of what was going on… with repo 105 could have lead risk management to conclude that the real leverage was much higher but was being shifted off balance sheet to hide the real risk.

    7. Weakness is not with sensitivity analysis as a method but limitation is with our understanding. If we presume random walk of stock prices/returns, when they are not, it’s our limitation. We know that return time series doesn’t follow normal distribution but still, if we cant find solution of closed form PDE for options (unless we assume normal or chi square distribution), its our limitation. If we can’t model correctly interaction of variables, and assume they are “independent", we are to blame. Finally, if we can’t have a "closed to reality still a hypothetical situation to convince higher management", it’s our limitation, and not of SA.

    Please understand that application of SA in pure science and social sciences (like banking and finance) is quite different and assumptions there may not work here. One has to test all assumptions before applying it and, with due respect to all, I know a handful of us really do so and blame SA for unexpected results.”

    “Could You Name Three Best Patterns in Trading [in general]?

    1. There are no best patterns. There are a few good patterns, which do well and also at times don’t do well. I don’t think there is any "maximize" or "optimize" criteria here…

    2. It’s a really good question. My only apprehension is that when you say pattern are you talking about the ones which people discuss all the time or some particular patterns which is more or less proprietary. I am assuming here you are talking about general patterns. I can suggest you a paper from Andrew Lo "Foundations of Technical Analysis: Computational Algorithms, Statistical Inference, and Empirical Implementation "…you can search it over net. I think it will help you.

    3. Head n Shoulder, PopGun, 123 top/bottom, triangles.

    4. Just try the following combinations; multiple exponential moving averages (14p, 30p,100p,150p and 200p) + 2.5 sd bollinger bands + ichimoku pattern… The above combinations should be simultaneously applied to 5 minutes chart, 30 minutes chart, 1 hour chart, daily chart and weekly chart. Most of the time, it provides perfect direction of the market, and I m sure you'll definitely appreciate after using this.

    5. Parametric models depend on assumption of underlying distribution. If that assumption is correct, parametric models are far better than non-parametric ones. However, in the real world, the underlying distribution is generally not known and often is non-stationary. Then the non-parametric models, such as AI, can give much better results. The problem is that, like any other tool, AI, if misused, can give grossly erroneous results.

    The traditional parametric models, such as regression, also have that problem. Misuse it and you will get spurious correlations. For example, I can prove with simple regression that the height of the tree in my backyard predicts US GDP - both are increasing with time. With black-box AI, where no-linear relationships are harder to track, such ridiculous results are harder to spot. That is why theoretical understanding and out-of-box testing are so important.

    6. Sometimes the simpler the better when it comes to short term stock picking. I have found these 3 basic steps lead to a statistically valid edge for locating stocks ready for gains within the next 5 days

    7. I am working now for 20 years in industrial data mining and machine learning. If you recognize features in technical processes and use them for decisions and predictions, the predictions do not change the values (of a temperature,…) , but in finance they do.

    What if we all use the same decision patterns?

    So, provocatively speaking: search patterns that help you to change market behavior in your segment and change them… I relate to Aaron Brown's book "The Poker Face of Wall street" - when your playing becomes predictable, you lose - change your strategies frequently and unpredictable. In other words, when patterns become think-stoppers… This does not mean that the help of evolutionary programming approaches with trading patterns are completely senseless.

    8. …I do believe that the only reason these indicators work is a self-fulfilling prophecy; there are markets where these indicators work better (i.e. not as bad) just like you mentioned, and generally these are the markets that are more liquid and are thought of as followed by a large nr of traders of various levels of sophistication. I think currency market is a prime example. I did a silly back testing of just one of these indicators when I was at school on a daily series of G10 ccies, 5 stock indices (snp, tsx, dax, cac, ftse) and a short list of major stocks - yes, MSFT and IBM were included. As far as I remember the worst performance was for the individual stocks and the best for the ccies.

    Another point worth mentioning is that there are some technical indicators that are based on discretization of models that are used in financial economics as good theoretical models for assets, such as mean-reversion for example. Momentum is another one of these indicators that can be claimed by both chartists and quants. That factor has a good theoretical underpinning and has been shown to be additive in a number of studies and across various assets and geographies.

    9. The three best Patterns in Trading: Trend Following, Contrarian, and Range Trading. If you think a little more about it, this can regroup into two groups; Trend following and Contrarian, which can be regrouped into one single: Trend Following. (A contrarian trader tries to catch a new trend right before it begins) The rest is details: Underlying market psychology covered by the system, associated entry management, position management, risk management, exit management… the devil in the details.

    10. With respect to the comments above referring to high frequency trading how much of HFT is really dependant on customer flow and or co-location of servers. I always found that on daily data it was the simplest systems that worked best. It is apparent right now that the fad de jour on wall st is high frequency trading in the same manner that repackaging mortgages was the fad de jour until it caused the financial melt-down. So given that everyone and there grandma is now diving into HFT with abandon and that it is hard to get a handle on the actual leverage deployed at what point is it all going to blow up in everyone’s face. That being said there probably is some merit to technical systems such as moving average crossovers, RSI the identifications of heads and shoulders and flags.

    11. Honestly, I don't know any grandmas that know about latency and stock processes, but I do get your point; there are plenty of banks/hf/institutional investment managers who are diving into the area; the supply of quants for these roles is only bounded by the supply of physics/math PhD’s… if anything, they at least will provide liquidity to the market and will probably add 0 alpha over the long term as a group; it's a zero-sum game, right?”

    Note: I have excerpted and edited these comments from the cited forum.


Resources & Links