frEaKaNaLyTIcS

Loss minimization model in forex trading

2009-09-26T01:58:00.005-04:00

The mathematical model below describes how loss can be minimized while trading in the forex market. Assuming we trade USD/EURO and give an order to sell hoping that there would be an increase of 10 pips. However, if the there is an unexpected decrease in pips, open up the 1 minute chart. Having studied the one minute chart carefully, I believe there is always a chance that there will be a 10 pip increase during a fall of 80 pips or less. In some cases there is a fall of more than 80 pips without a change in the graph pattern. However, those can ignored as unexpected anomalies. So, our goal is to make up for the lost money in the 10 pip increase in the middle of a sharp fall.

Lots	Pips	Profit/Loss	Money Spent	Sum of Pips	Sum of Lots	Lots to Equalize(+10 Pips)
1	-20	$ (20.00)	$ 75.00	-20	1	1
2	-20	$ (80.00)	$ 225.00	-40	3	5
5	-20	$ (240.00)	$ 600.00	-60	8	16
16	-20	$ (720.00)	$ 1,800.00	-80	24	48
48	-20	$ (2,160.00)	$ 5,400.00	-100	72	144
144	-20	$ (6,480.00)	$ 16,200.00	-120	216	432
432	10	$ -	$ 48,600.00	-110	648	-648

In this model, we buy 1 lot assuming that there would be a 10 pip increase. However, there is a sharp fall. We wait for a 20 pip decrease and then sell another lot hoping that a 10 pip increase would neutralize our loss. Similarly, when there is another 20 pip fall, we sell another 2 lots hoping that a 10 pip increase would neutralize our loss. The process continues and it is highly unlikely that there is never a 10 pip increase across a fall of 120 pips. Hence, even if things don't go as planned, there is always a way to get your money back!

Lots	Pips	Profit/Loss	Money Spent	Sum of Pips	Sum of Lots	Lots to Equalize(+10 Pips)
1	-20	=SUM(B1,B2)	=(F2*75)	=SUM(B2)	=SUM(A2)	=-(C2/10)-F2
2	-20	*=((F3B3)+C2)**	=(F3*75)	=SUM(E2,B3)	=SUM(F2,A3)	=-(C3/10)-F3
5	-20	*=((F4B4)+C3)**	=(F4*75)	=SUM(E3,B4)	=SUM(F3,A4)	=-(C4/10)-F4
16	-20	*=((F5B5)+C4)**	=(F5*75)	=SUM(E4,B5)	=SUM(F4,A5)	=-(C5/10)-F5
48	-20	*=((F6B6)+C5)**	=(F6*75)	=SUM(E5,B6)	=SUM(F5,A6)	=-(C6/10)-F6
144	-20	*=((F7B7)+C6)**	=(F7*75)	=SUM(E6,B7)	=SUM(F6,A7)	=-(C7/10)-F7
432	10	*=((F8B8)+C7)**	=(F8*75)	=SUM(E7,B8)	=SUM(F7,A8)	=-(C8/10)-F8

COLUMN NAME	EXPLANATION
Lots	Number of Lots purchased each time
Pips	Number of Pips increased/decreased after purchasing lots
Total Profit/Loss	Total profit or loss incurred after the most recent purchase (amount within brackets indicate loss)
Money Spent	Amount spent to purchase the Lots each time
Pips Inc/Dec	Difference in Pip count since we started buying
Sum of Lots	Total number of Lots purchased so far
Lots Req. to Equalize(+10 Pips)	Number of Lots that are to be purchased to neutralize loss if we expect an increase of 10 pips

The IPL Economy

2009-04-18T00:53:00.027-04:00

When the British came to the subcontinent, there were many kingdoms - each with a capital town, a distinctly different language, a different and unique culture and religion. The British in a way helped unite these forces together to create India. Thank God, we didn’t have money launderers (stupid financial institutions which crash every hundred years) in the 18th century; the British were economically quite sound to handle large colonies for 300 years and India was born. If the British had given the subcontinent a miss, we would have had 20 to 30 countries in the subcontinent and all of them would have been constantly fighting with each other over water and land. We would have seen the creation of several Islamic republics, Hindu majority states and possibly a Sikh country; most of them having just a few things in common: poverty, illiteracy, etc The region that now boasts to be the largest democracy in this world could have played host to several autocratic and military regimes. Good Lord! I love Britain. They saved the world from possibly several North Koreas and Zimbabwes and instead presented the world with a flourishing democracy!

The British stayed long enough and were cruel enough to unite the people of the Indian sub-continent. For the first time, people in this region started to look beyond caste, religion and linguistic divides to call themselves 'Indians'. The India that we have now is by far the most diverse country in this world. Unity in diversity - India's beauty and strength.

This unification of 1 billion people also created the market and the workforce that an emerging economy needs. I know that a large percentage of that population cannot read and write, is below poverty level, not efficient workforce, etc. However, let’s assume we have just 1 in 3 Indians who are well educated, skilled and have the buying power - that’s 300 million people which is the total population of the United States. The companies around the world are investing in India because of our huge market that is growing every day.

A comparison of per-capita coke consumption in US and India:

* Based on U.S. 8 fluid ounces of a finished beverage

(Source:http://www.thecoca-colacompany.com/ourcompany/ar/percapitaconsumption.html)

This graph shows the per capita coke consumption in US has not seen much increase in the last 10 years. Though, India has doubled its tally, it's still way behind US. Imagine a country with a population of 1 billion having a per-capita coke consumption that is 60 times smaller than that of America. Now, that is a market to be captured! If I were Indra Nooyi, I would buy an IPL franchise the next time BCCI decides to expand the number of IPL teams. I'll tell you why...

For those who have no idea what the IPL is, wikipedia: "The Indian Premier League (also known as the "DLF Indian Premier League" and often abbreviated as IPL), is a Twenty20cricket competition created by the Board of Control for Cricket in India (BCCI) and chaired by the Chairman & Commissioner IPL, BCCI Vice President Lalit Modi. It is now the most watched domestic Twenty20 competition in the world. The first season of the Indian Premier League commenced on 18 April 2008, and ended on 1 June 2008 with the victory of the Rajasthan Royals against Chennai Super Kings in the final at the DY Patil Stadium, Navi Mumbai."

To add more, it is a franchise based club cricket competition based on the English premier League and the American franchise-based competitions. The loyal following every EPL club has earned over 100 years has turned the EPL into a huge industry. A club like Manchester united commands a great amount of fan following even in Asia.

England

being such a small country without the kind of diversity that

India

has produced such a successful league. Imagine the kind of market, a diverse country like

India

proposes. When Liverpool and Manchester United were established, there was no immediate fan base for these clubs apart from the people living in and around the region. However, look at the kind of fan base the Kolkata Knight Riders or the Chennai Super Kings can generate. With each region in India having such a uniquely distinct cultural identity, and each culture trying to constantly establish their supremacy in a virtual culture rankings, the rivalry and fan base for these clubs is going to grow manifold in the years to come. Cricket being the most popular sport in

India

, the IPL has to be the biggest marketing tool in the future.

Now coming back to why The Coca-cola Company should invest in an IPL franchise, it is quite apparent that the per-capita coke consumption of an average American has reached a saturation level and it is very difficult to go beyond the current level. Hence, it makes more sense to approach emerging markets like India which have a tremendous potential for growth. The Coca-cola Company can almost double it's profits through coke sales if it can replicate even one-third of the American coke consumption in India. For that, they need to market their product very well. As I had mentioned before, with the IPL being the most appealing marketing tool for the future, I see a lot of firms investing in it. With more investment, the IPL is only going to grow bigger. The other big advantage that IPL holds over EPL is that, Indians living abroad are more than half of the total population of England! So, the market for IPL expands beyond Indian borders instantly, unlike the EPL which took several years to capture markets in Asia and Middle East.

I wouldn't be surprised if after a few years, bars in England screen IPL matches (if it does not clash with a English premier league match of course, not in England!)

Disappearing Unknowns

2009-04-07T15:09:00.006-04:00

Player	Span	Mat	Runs	HS	Bat Av	100s	Wkts	BBI	Bowl Av
MA Noble (Aus)	1898-1909	42	1997	133	30.25	1	121	7/17	25.00
*WW Armstrong (Aus)	1902-1921	50	2863	159*	38.68	6	87	6/35	33.59
JV Saunders (Aus)	1902-1908	14	39	11*	2.29	0	79	7/34	22.73
*FE Woolley (Eng)	1909-1934	64	3283	154	36.07	5	83	7/76	33.91
MW Tate (Eng)	1924-1935	39	1198	100*	25.48	1	155	6/42	26.16
DVP Wright (Eng)	1938-1951	34	289	45	11.11	0	108	7/105	39.11
*WA Johnston (Aus)	1947-1955	40	273	29	11.37	0	160	6/44	23.91
JH Wardle (Eng)	1948-1957	28	653	66	19.78	0	102	7/36	20.39
*JR Reid (NZ)	1949-1965	58	3428	142	33.28	6	85	6/60	33.35
*GS Sobers (WI)	1954-1974	93	8032	365*	57.78	26	235	6/73	34.03
DL Underwood (Eng)	1966-1982	86	937	45*	11.56	0	297	8/51	25.83
*AW Greig (Eng)	1972-1977	58	3599	148	40.43	8	141	8/86	32.20
B Yardley (Aus)	1978-1983	33	978	74	19.56	0	126	7/98	31.63

(Source: http://stats.cricinfo.com/statsguru/engine/stats/index.html?bowling_pacespin=3;class=1;filter=advanced;orderby=start;qualmin1=75;qualval1=wickets;template=results;type=allround)

The table shows the list of bowlers classified as 'mixture/unknown' by cricinfo. The results were filtered to view such bowlers who have taken more than 75 test wickets and were sorted by date. Some of them are batsmen who just rolled their arm over as part timers and hence, they cannot be considered professional bowlers. A '*' has been marked next to their names to differentiate them from players who were in the team predominantly for their bowling.

It is quite apparent that there is no bowler in that category after 1983. Bruce Yardley of Australia was the last player to be categorized as 'unknown' and take more than 75 wickets. He retired from international cricket before I was born! What does that tell? Cricket has evolved as a professional sport in the last three decades and there is no place for the unknowns anymore.

Industrial Engineering in Health Care

2009-04-07T13:01:00.013-04:00

Most engineering disciplines apply skills to very specific areas, whereas industrial engineering is applied in virtually every industry. Hence, Industrial engineering is about choices.

Industrial Engineering, in my perspective, is all about making things better and in Health Care we are more often dealing with 'lives' rather than 'things'. Imagine the kind of satisfaction you would have when you have played a role in saving a child's life rather than bettering the manufacturing of a car.

Increasing requirements for cost control, operational efficiencies and the emergence of large health care systems are key factors in the health care industry's growing need for Industrial Engineering to improve performance.

Equating health care problems with typical industrial engineering problems

Industrial Engineers graduates often wonder what they would be doing in a hospital. Here are a few comparisons that could help bridge the gap between typical industrial engineering problems in other industries and health care issues:

Patient flow - supply chain
Surgical wait list - better scheduling
Infectious diseases - logistics
Health human resources – forecasting

Why the health care industry needs industrial engineers?

Some of the glaring problems in the health care industry in US are given below and who better than industrial engineers can handle these complex and unstructured problems:

Every $1 out of $6 of the US GDP (or 16%) goes to healthcare – largest component of federal and State budget
Other western European nations spend less than 10% on healthcare.
In 2006, $2.1 trillion was spent on healthcare. That is $7,026 per capita, which is a 6.7% increase over 2005.
At this kind of growth, it is projected that $4.1 trillion will be spent in healthcare in 2016 (1/5 of GDP).
5% of the US population account for 50% of the healthcare expenditures.
Prescribed medication expenditures almost doubled from 12% to 21% in ~10 years.
The largest expenditure is on inpatients (over 30%).
The second largest is on ambulatory care (over 20%).
Chronic diseases (heart, cancer, trauma-related disorders, mental health, and pulmonary) account for a majority of the expenditures.
Medical errors accounted for 44,000 avoidable deaths in hospitals a year.
Americans are less healthy: 33% obesity rate and high rate of chronic diseases.

(Source: Access & Cost Trends at the Agency for Healthcare Research & Quality (AHRQ))

Hospital-acquired infections:

2 million NSI per year, $3,000/episode
8.7 million additional hospital days, 20,000 deaths / year

Adverse drug events:

770,000 to 2 million per year
$4.2 billion annually

Medical errors and caregiver-induced injury:

98,000 deaths / year, 770,000 - 2 million patient injuries / year
$17 - $29 billion dollars
More US deaths / year than for traffic accidents, breast cancer, & AIDS

(Source: Society for Health Systems)

Industrial Engineers can provide policy makers with informed recommendations, surveys, simulations, forecasting models, and other operations research tools to solve these difficult problems.

Challenges of applying industrial engineering in health care

Analysis:

Lack of relevant data and/or scarcity of valid data
Weak structure of the system
Qualitative parts tend to dominate

Modeling:

Difficulty of reaching the right level of abstraction
Difficulty defining goals and objectives

Applying:

Delivery care planning is often a matter of politics (not a question of calculation)
Hospital administrators: Limited exposure to quantitative methods

A new way of looking at the healthcare system's stakeholders is given below. Looking at it, it is no surprise that it's difficult to make decisions in a hospital!

(Picture Source: "New Challenges for Operations Research Applications in Health Care" by Professor Mike Carter)

Creative ideas for industrial engineering applications in health care

Applications of industrial engineering often involve a lot of creativity. From my experience, I believe creativity is the most important quality an industrial engineer should have. You might be wondering that this should be because of the importance of coming up with creative solutions. You might be partially correct. However, in my opinion, creativity is more important because it helps identifying problems.

I am describing two very simple and creative problems in this section. The first one is identifying the most important factors that affect the waiting times in hospitals. A regression model can be built with the emergency room waiting time as the dependent variable and several factors affecting waiting times as independent variables. The factors that affect waiting times could be number of staff at counter, number of pharmacies, number of doctors, number of nurses, number of computers, efficiency of hospital staff (number of patients attended per unit time), number of errors in patient appointment timings, etc. The data could be collected from various hospitals and data collected from each hospital would be an observation in the regression model. To get the best results, we need more observations and hence, data from a large number of hospitals.

Hospital	ER Wait time (min)	staff at counter	pharmacies	doctors	nurses	computers	patients/hour	appointment errors/month
Emory	20	5	6	22	25	10	3	2
Piedmont	30	10	14	23	30	15	4	3
Grady	50	4	2	15	20	5	1	5
Northside	18	3	5	18	40	20	2	1

(The data in this table is completely fictional)

The table above has fictional data and shows the different factors that affect the emergency room wait times. To build a regression model we have to include several other possible factors and also need to have large number of observations from different hospitals. The data was similarly simulated for hundred observations using random number generation and a regression analysis was done.

	Coefficients	Standard Error	P-value
Intercept	30.9831	8.399657	0.000383
staff at counter	-0.043821	0.408822	0.114876
pharmacies	-0.261887	0.239263	0.5766
Doctors	-0.60296	0.235891	0.120015
Nurses	-0.059476	0.136625	0.36436
Computers	-0.22109	0.181268	0.225725
patients/hour	-0.08138	0.86755	0.925474
appointment errors/month	0.21664	0.631001	0.732148

(The p-values have been altered from the original result to explain the process better)

The regression model above shows the p-values of each of the factors considered. From the results, the factors with very low p-values are statistically significant and can be considered for further analysis. The level of impact of each factor can be obtained from the coefficients.

Statistical data mining tools like regression tree and k-means clustering can be used group the hospitals into specified number segments based on the factors mentioned in the model and even demographic factors of the location of the hospital. This would help us classify hospitals and approach problems of each group in different methods. Advanced clustering algorithms like chi-squared automatic interaction detection clustering can group the hospitals into an optimized number of clusters.

Dogs in the name of God

2009-04-05T03:53:00.005-04:00

Long time back when dogs were still wolves.. a stupid man stood, observed and decided to use his sixth sense to save the human race from self destruction. The stupid man went to his tribe and said he saw a guy called god who could walk on water and more importantly had the powers to grant any wish. However, the grace of this guy called god had to be earned by following certain rules. These set of rules which were primarily intended to spread peace and love grew with every generation and branched of into religions. It did not end there. Just as some of us like Chinese food, some Italian and some Indian, primitive men had their own preferences. Some wanted their god to fly, while others wanted a god with healing powers... just as we choose superheroes with different powers in video games. As if that was not enough, they even argued over the looks of god. Some wanted him to look rich and well groomed; some wanted him to be rugged and bearded, etc. There was no harmony of opinion and people decided to split. Each group had a god of its liking: Allahs, Jesus Christs, Rams and Buddhas.

Thousands of years passed, man evolved, god and religion also evolved with him to take several forms. However, peace and love still evades humanity. The stupid man who created god should have realized that humans are intrinsically narcissist. We are self obsessed and programmed to believe that we are second to none and hence, this constant infighting. We constantly look for new avenues to fight amongst ourselves. The greatest irony is that an entity that was created to unite us has become the latest avenue for fighting. Gone are those days when god protected man, its man who is protecting god now, that too from his own peers! The third world war would not be fought for land, money or oil. It would be Ram vs Jesus, Abraham vs Allah, etc

Recently a Hindutva moral brigade attacked women in a pub in Mangalore, India. Apparently the mob was motivated by Sri Ram Sene, one of the many right-wing Hindu organizations which are self proclaimed protectors of the so called Hindu culture of India. It is painful to watch such incidents, especially when people do it in the name of god. Thousands of young minds around the world are indoctrinated in terror camps so easily, all in the name of god. A bunch of young kids carrying AK47 rifles brought India to a standstill on 26/11. They killed many and laid their lives down, all for god.

So, do we need god and religion? I don't think so. For god's sake, we can do without all this mess. It's ruining our lives.

I am not an atheist. I do believe in God (the first time I'm using a capital G in God, because I'm referring to the real God!), not the ones created by man though. I believe that there too many mysteries to be resolved by science and until the last one is resolved, I will continue to believe in God. The last mystery would probably be God himself!

No Foe No Votes!

2009-04-04T21:42:00.002-04:00

Bomb Blasts, carnage on live TV, celebrating martyrs, candle light vigils, talk shows detesting politicians, talk about people's resilience, patriotic fervor, call for peace from the west, Indian and Pakistani governments bashing and blaming each other, oppositions demanding more bashing, ELECTION, the party that does most bashing wins, peace, public's memory loss... another bomb blast...

This pattern has been repeating itself for more than a decade now and I don't see it ending anytime soon. So, can India and Pakistan ever be friends? If that happens, what would the politicians talk about in election rallies that are flocked by people who turn up for free food and hate speech?- Infrastructure development? Nuclear deal? Lunar missions? No....

After all you need a villain to become a hero!

Evolution Captured, Capsuled

2009-03-25T22:40:00.006-04:00

The beauty of evolution is that its massive impact is masked by its virtually impalpable actualization. Rarely can we pause and observe the process of evolution because it's so subtle and imperceptible. However, the cricketer is a species that is going through a very discernible and at times disturbing transition right now. Cricketers these days are born with a mutant gene; a gene that used to tell them to go forward and defend a good length delivery is asking them to heave it over long on for a six! Of course, this mutation is not so prevalent in a sub-species called bowlers, but I thought it was not worth talking about a group that is going through extinction! Twenty20 is the UV radiation that is causing this mutation. Before you make an assumption let me clarify, this is not a usual T20 hate article by a new breed of hypocritical, traditional and self-proclaimed purists. I love T20 and I am a Chennai Super Kings fan!

Gone are those days when 280 was a very respectable score in ODIs and I'm not talking about too long ago. Since, the T20 world cup in South Africa, 320 on the board has become a minimum requisite to challenge the opposition on a good wicket. Batsmen have started hitting sixes such ease that it does not anymore deserve a rousing reception. The high ODI scores can be largely attributed to the increase in number of sixes, and the number of sixes can be attributed to the inception of T20.

An analysis of ODIs 2 years before and after T20 World Cup in South Africa:

Team	Matches	6s	6s per match
Australia	143	464	3.244755245
India	142	379	2.669014085
New Zealand	111	362	3.261261261
Pakistan	116	353	3.043103448
South Africa	111	346	3.117117117
West Indies	115	342	2.973913043
England	117	263	2.247863248
Sri Lanka	130	214	1.646153846
Bangladesh	102	180	1.764705882

Table 1: Number of 6s per match by team from 01 Jan 2003 to 25 Sep 2007 (Source:http://stats.cricinfo.com/statsguru/engine/stats/index.html?class=2;filter=advanced;groupby=team;orderby=runs;spanmax1=25+Sep+2007;spanmin1=01+Jan+2003;spanval1=span;template=results;type=batting)

Team	Matches	6s	6s per match
India	51	205	4.019607843
New Zealand	39	173	4.435897436
West Indies	30	132	4.4
Australia	38	102	2.684210526
Pakistan	34	101	2.970588235
Sri Lanka	43	86	2
England	28	74	2.642857143
South Africa	30	74	2.466666667
Bangladesh	35	49	1.4

Table 2: Number of 6s per match by team from 25 Sep 2007 to 22 Mar 2009 (Source:http://stats.cricinfo.com/statsguru/engine/stats/index.html?class=2;filter=advanced;groupby=team;orderby=sixes;spanmin2=25+Sep+2007;spanval2=span;template=results;type=batting)

Team	6s per match (2003 to T20 WC)	6s per match post T20 WC	Difference	W/L (post T20 WC)	W/L (2003 to T20 WC)	Difference
Australia	3.244755245	2.684210526	-0.560544718	1.83	3.72	-1.89
Bangladesh	1.764705882	1.4	-0.364705882	0.29	0.48	-0.19
England	2.247863248	2.642857143	0.394993895	0.71	0.92	-0.21
India	2.669014085	4.019607843	1.350593759	1.93	1.19	0.74
New Zealand	3.261261261	4.435897436	1.174636175	1.46	1.25	0.21
Pakistan	3.043103448	2.970588235	-0.072515213	2.09	1.33	0.76
South Africa	3.117117117	2.466666667	-0.65045045	2.62	1.66	0.96
Sri Lanka	1.646153846	2	0.353846154	1.05	1.45	-0.4
West Indies	2.973913043	4.4	1.426086957	0.58	0.91	-0.33

It is quite apparent that India's performance has improved significantly post T20 WC and it also shows a huge increase in number of sixes per match. Indian players have also been exposed to the IPL more than other players and hence the heavier T20 influence . Pakistan has also shown a good improvement after T20 WC but does not show an increase in sixes per match. Since most of those matches have been against weak oppositions Pakistan's case can be ignored as an anomaly. England players did not take part in IPL and that probably explains the marginal increase in sixes per match in the past two years and as expected their overall performance has also deteriorated. Though the six hitting phenomenon triggered by T2oWc and IPL is only one of several factors influencing increased ODI batting standards, it's quite clear that it has has a major impact on the performance of teams.

The same sets of players have suddenly unleashed the devil within them in the last two years by hitting towering sixes in ODIs. This has been possible because T20 has given them a good platform to practice six hitting in match conditions. My chauffer, Mr.Ekambaram has been driving cars for a profession for the past 25 years and yet he would struggle to race through a narrow street as I can. That is because; during the past 25 years he has never practiced driving like crazy through narrow streets. Who knows, if Geoff Boycott was exposed to IPl at the age of 20, he might have ended up as a pinch hitter like Yusuf Pathan. However, the downside is that without exposure to IPL youngsters like Rohit Sharma and Robin Uthappa might evolve as better test batsmen, arguably though. Will see you next time with more analysis!

Cheers