Quriosity
Monthly Newsletter of Quantinuum@SIMSR
January 2018 | Volume 09, Issue 01
GOOGLE ANALYTICS
The Turing Test 1
Quriosity | Vol. 09, Issue 01 | January 2018
CONTENTS Curabitur varius lobortis lectus, comments to: quis pharetra eros newsletter.quantinuum@gmail.com dictum sit amet non volutpat Editorial Team: Rupal Doshi
Please submit your articles & your
Amod Kulkarni Aditya Gupta Chaitanya Agarwal VVNS Anudeep Kapil Gupta Samoshri Mitra Khushbu Mehta
COVER STORY-
GOOGLE ANALYTICS
SUB ARTICLE-
SURVEY SAMPLING
QUANT GURU-
S.R. SRINIVASA VARADHAN
CURIOSITY UPDATE NEWS DIGEST QUANT FUN QUANT CONNECT
Designing Team: Shreyas Kulkarni Ashish Mahadik
www.quantinuumsimsr.com
1
Quriosity | Vol. 09, Issue 01 | January 2018
EDITOR’S NOTE
Welcome to the latest issue of Quriosity, the monthly newsletter of Quantinuum! Many current data analysis techniques are beyond the reach of most people. Obscure math and daunting algorithms have created a chasm for problem solvers and decision makers. Quriosity is trying to bridge these gaps by giving appropriate inputs to our students and readers who are the future managers. The objective of Quriosity is to publish up-to-date articles on data analytics, alongside relevant and insightful news. This way the magazine aspires to be vibrant, engaging and accessible, and at the same time integrative. The cover story by Tejal Jadhav is based on how the “GOOGLE ANALYTICS” works and all about its features and limitations. The article on“SURVEY SAMPLING” by Shalini Hirawat explains how this stage of market research is taking a transformational shift from qualitative to quantitative approach. We have an article on Quant Guru “S. R. Srinivasa Varadhan”, highlighting the contribution of the Indian mathematician & statistician in multidimensional difusion process. Any articles that you wish to submit, either individually or collaboratively, are much appreciated and will make a substantial contribution to the development and success of the magazine.
Thank you and Happy Reading! Editorial Team Quantinuum@SIMSR
2
www.quantinuumsimsr.com
Quriosity | Vol. 09, Issue 01 | January 2018
GOOGLE ANALYTICS
-By Tejal Jadhav
Google analytics, the word itself explains that it is web analytics service provided by Google. So now the question comes, what is web analytics? Web analytics is the process of analyzing the behavior of visitors to a Website. It also helps in measuring website traffic, market research, and improving effectiveness of website. Following are the Features of google analytics, 1. Review online campaign traffic: Helps in analyzing performance of campaign in terms of traffic visiting website 2. Tracking goals: Website conversions & user activity: Which can be done by evaluating following a. Destination b. Duration c. Pages/Screens Per Session d. Event or action example: click 3. Audience reports: These reports give you insights of demographics and geo-location of interests and key browsing behaviors 4. Flow visualization: Here it becomes easier to understand how the consumers are getting hooked to website content and what is the end destination of their journey, do they get converted or not? 5. Google Website Optimizer: This is the part of google analytics, which can handle multivariate experiments 6. Cohort analysis: This feature helps understand the behavior of component groups of users apart from your user population
2. Data storage Data of past 25 months is available without any charge, but data availability beyond 25 months is chargeable 3. Tracking limitation 4. Blocked cookies The largest potential impact on data accuracy comes from users deleting or blocking Google Analytics cookies. Without cookies being set, Google Analytics cannot collect data. Any individual web user can block or delete cookies resulting in the data loss of those visits for Google Analytics users 5. Time calculation Can not calculate time spend on the last page of a session hence data available for analysis is inadequate Comparison with other web analytics software:
Conclusion Google analytics is a great device but it has got some limitations too, it cannot be used if you are managing a big website for example, an e-commerce website, which has great amount of traffic then it is advisable to use the premium version of google analytics as it comes with additional features and limits. Though, the free version of google analytics may have some limitations, it is still best available free web analytics tool for small and medium business website. Sources: 1.http://www.digitalgenx.com/11limitations-of-google-analyticstracking/ 2.https://en.wikipedia.org/wiki/ Google_Analytics
Though google analytics provide us various features, but it also has some limitations, which are as follows, 1. E-commerce transaction limit Only 50,000 e-commerce transactions per day are recorded, transactions beyond this limit are ignored
3
www.quantinuumsimsr.com
Quriosity | Vol. 09, Issue 01 | January 2018
Survey Sampling Statistics deals with a large number of figures and not just a single figure. It includes mathematical analysis with help of quantified model to review, analyse and draw conclusions from the data gathered. Pool of all the items considered for analysis is termed as population. Population may be finite or infinite and it is difficult to test the entire population to draw any conclusion as it involves time, energy and money. Therefore, a subset of the population, known as sample, is used, on which statistical tools are applied to draw the inferences with respect to the entire population. The process of selecting representatives (sample) from all the elements of the population is called Sampling. It helps in reducing cost and/ or time and/or amount of work involved in examining the entire population. Survey sampling includes the process of sampling to conduct a survey, which solicits information from people. It most often involves questionnaires to measure the characteristics, attitudes or perceptions of people. Surveys can be public survey, such as opinion polls during elections, as well private which are used extensively by market researchers. Surveys with a low response rate, where response rate refers to the proportion of all people who are selected who complete the survey, can lose validity of any conclusions drawn from statistical analysis. Many researchers are of the opinion that Personal Interviews is the best mean to survey people, where interviewer solicits information from a respondent by asking prepared questions giving it a personal touch. Here, response rate is comparatively higher with fewer incorrect responses as the misunderstandings are instantly cleared. However, along with higher costs involved, there can exist biases which may question the validity of the sample to give true picture of the population. Biases can be non-response bias, response bias, selection bias, participation bias, coverage bias, etc. Another mode of survey can be through telephonic interview, which is less
4
expensive but is less personal and has a lower expected response rate. SelfAdministered Surveys is an inexpensive method of surveying where questionnaire is sent to a sample of people. But, it usually has low response rate and a relatively high number of incorrect responses due to respondents misunderstanding some questions. Sampling of the data collected through one or more of the aforementioned methods can be done based on either probability or non-probability method. Probability sampling provides a scientific technique of drawing samples from the population, where each element has a defined pre-assigned probability of being chosen in the sample. Within probability sampling, there are specialized techniques such as random sampling, stratified sampling, cluster sampling and systematics sampling that improve the precision or efficiency of the sampling process by reducing sampling error. A probability-based survey sample is created by constructing a list of the target population, called the sampling frame for experiment. In Random Sampling, every element of the population has an equal chance of being included in sample and thus, such method is suitable where population is homogenous. Stratified Random Sampling is used when population is heterogeneous. Here, the elements
-By Shalini Hirawat
are categorized into groups (strata) that are homogeneous and then the sample is drawn each stratum. In Cluster Sampling, the population is divided into some acceptable sub-divisions known as ‘clusters’ and then samples are drawn from such clusters to conduct survey. In Systematic Sampling, the sample is selected from the population on the basis of some systematic arrangement of the entire population. Many of the surveys are not based on probability based sampling, but rather on finding a suitable collection of respondents to complete the survey. Such methodology is known as non-probability sampling. Such technique will always produce larger sampling errors due to non-random nature of selection process. Within Non-probability sampling technique, purposive or judgement sampling, quota sampling and convenience sampling are covered. Purposive sampling involves selecting elements from the population to comprise a sample such as they specifically meet some prescribed purpose of specific attributes of interest that address the purpose of a particular research problem under investigation. In Quota sampling, the data is stratified into small units called quotas prior statistical analysis. And Convenience sampling is highly unsystematic, careless and accidental www.quantinuumsimsr.com
Quriosity | Vol. 09, Issue 01 | January 2018 one because the samples are drawn based on convenience of the researcher. Decision with respect to data collection source and sampling method to be used plays a very crucial role. There is no defined sampling technique application and thus, it always depends upon the purpose, quality and effectiveness of data and the researcher. Sometimes in reality it so happens that, a study requires a mix of above mentioned techniques rather than a specific technique. Such technique is known as ‘Mixed sampling’. Only appropriate technique will not result in required data analysis and desired results. nstead, researcher should be very careful during sample designing, determining appropriate sample size and data collection method combined with appropriate statistical sampling technique to arrive at the desired inferences about the population. In Random Sampling, every element of the population has an equal chance of being included in sample and thus, such method is suitable where population is homogenous. Stratified Random Sampling is used when population is heterogeneous. Here, the elements categorized into groups (strata) that are homogeneous and then the sample is drawn each stratum. In Cluster Sampling, the population is divided into some acceptable sub-divisions known as ‘clusters’ and then samples are drawn from such clusters to conduct survey. Sources: a. Statistics for Management and Economics – Gerald Keller b. Research Methodology – Dr. Prasant Sarangi c. Business Statistics – J.K. Sharma d. Internet
5
Quriosity| November 2017
Quant Guru:
S.R. Srinivasa Varadhan
- Dropad Saxena
Sathamangalam Ranga Iyengar Srinivasa Varadhan was born in Chennai on 2nd January 1940. He completed his undergraduate education at the Presidency College, Chennai and received his B.Sc. Honours degree in Statistics from the University of Madras in 1959. He then joined the Indian Statistical Institute at Kolkata, where he earned his Ph.D. degree in 1963 for his research work on infinitely divisible probability measures on an infinite dimensional Hilbert space. He then migrated to the USA and joined the Courant Institute of Mathematical Sciences, New York University, where he is currently Professor of Mathematics and Frank J. Gould Professor of Science. Though motivated by problems from mathematical physics and the theory of partial differential equations, Professor Varadhan is essentially a probabilist. One of the highlights of Professor Varadhan’s scientific contributions is the construction of multidimensional diffusion processes based on the martingale method. This work, done jointly with D. Stroock, is still the most powerful technique in handling weak solutions of diffusion equations with non-smooth coefficients in both finite and infinite dimensions. In a long and
fruitful collaboration with M. Donsker, he founded a grand theory of large deviations in a very general setting. This theory, considered to be a mathematical tour de force, is much more subtle than the classical theory of limit laws in probability such as the law of large numbers and the central limit theorems. It combines probability theory with convex analysis, nonlinear programming, functional analysis and partial differential equations. The large deviation principle enunciated in this theory has wide ranging applications from actuarial sciences to mathematical physics. It led to his settling Pekar’s polaron conjecture and to doing impressive work in the area of hydrodynamic scaling limits, earning him the title of Grand Master of Large Deviations. Prof. Varadhan has held visiting positions in a number of esteemed institutions. He has received the Alfred P. Sloan Fellowship and the Guggenheim Fellowship. He won the Birkhoff Prize (1994), the Margaret and Herman Sokol Award of the Faculty of Arts and Sciences, New York University (1995), the Leroy Steele Prize of the American Mathematical Society (1996) and the Abel Prize of the Norwegian Academy of Science and Letters (2007). The Government of India has honoured him with the Padma Bhushan. Professor Varadhan has been elected fellow of several scientific academies such as the American Academy of Arts and Sciences (1988), the Third World Academy of Sciences (1988), the National Academy of Sciences, USA (1995), Institute of Mathematical Statistics, USA (1991), the Royal Society, London (1998),the Indian Academy of Sciences (2004) and the Indian National Science Academy (2008). Source: •Ramachandran, R. (7–20 April 2007). “Science of chance”. Frontline. India. Archived from the original on 11 December 2007. www.quantinuumsimsr.com
Quriosity | Vol. 09, Issue 01 | January 2018
CURIOSITY UPDATE Portable DNA Sequencer Searches for Life in Extraterrestrial-Like Conditions A tiny life-seeking instrument could be packed on a future mission to Mars or the icy moons of Enceladus or Europa. Its aim: to fnd microbes through a detective search of DNA and RNA and help us understand how life evolves on other worlds. The researchers recently tried out their techniques at a site on Axel Heiberg Island, a large Arctic island about 560 miles from the North Pole. Like Mars, Axel Heiberg is under a deep freeze most of the time. But it also hosts unique formations called cold perennial springs, which have roots in permafrost as deep as 1,970 feet. At sites measured by the researchers, the springs materialized in salty areas where microbes thrive. The new study showed that a portable DNA sequencing device called MiniON, manufactured by the United Kingdom’s Oxford Nanopore Technologies, works well in the feld. It detected microbes and also sequenced DNA from active microbes. While more tests need to be done, this frst step shows that searching for life on other worlds is defnitely possible. Cold, liquid, and salty sounds a lot like how conditions could be on Mars — particularly on features called recurring slope lineae. These are dark streaks that appear on the slopes of craters, particularly when the surrounding area is warm. Back in 2015, NASA announced that hydrated salts are present within RSLs, which is strong evidence of liquid, briny water fowing on the surface. (And water, of course, could mean life.) But later research has cast some doubt on that conclusion, with some researchers suggesting RSLs are piles of dirt tumbling down the slope, while others say atmospheric water is the source of the salts. It was pointed out by many researchers that there are potentially life-friendly sites all over the solar system. The Hubble Space Telescope spotted multiple water eruptions on Jupiter’s the icy moon Europa. The Cassini mission — which ended as planned last year — extensively mapped more than 100 geysers spouting from icy Enceladus, one of Saturn’s moons.
6
NEWS DIGEST Apex Heart Institute Becomes First Center In India To Commercially Install AI-Powered Vascular Robotic System In a historical development that can create new avenues in the field of interventional cardiology in India, Apex Heart Institute, Ahmedabad became the first location outside the US to have a commercial installation of the advanced AI-powered vascular robotic system, CorPath GRX. With the addition of this technology to its state-of-the-art cardiology facilities, Apex Heart Institute will become the first ‘Global Center of Excellence’ outside USA to train interventional cardiologists from India and across the world on robotic-assisted percutaneous coronary interventions (PCI). Japanese Researchers Create AI That Can Read Your Dreams A group of researchers from University of Kyoto in Japan have figured out a way where computers can visualize what you are thinking! Any guesses for how they have achieved it? Its—artificial intelligence. During the research which was carried over 10 months, three subjects were shown natural images such as photos of a bird or person, artificial geometric shapes and alphabetical letters for varying lengths of time. There are many potential applications, once the technology develops visible improvement. It can allow to draw pictures or make art simply by imagining something, it can visualize human dreams, hallucinations of psychiatric patients and much more. YES BANK’s Analytics Use Cases Receive Global Recognition The recently published Gartner report titled ‘Gartner Build Advanced Analytics and Data Science Capabilities: Lessons from the Gartner Excellence Awards, Kurt Schlegel, 28th November 2017’, has carried out YES BANK’s case study on successful deployment of data analytics for enhancing its business impact. India’s fifth largest private bank was among the three finalists shortlisted for the award. YES Insta, a first-of-its kind innovation in
the BFSI industry has also been featured in the case studies published by Gartner. InnerChef CEO Backs Use Of AI In F&B Industry, Says Food Delivery Platforms Don’t Share Enough Data With Cloud Kitchens After banking, retail and manufacturing, Artificial Intelligence has now also trickled down to the food and beverage industry. Earlier this week Indian cloud kitchen startup, InnerChef, announced its plans to expand its operations using data analytics and artificial intelligence in 2018.The startup that was founded in 2015, plans to increase its number of kitchens and add to its fresh food brands. Freshmenu too has been known for experimenting with AI in the recent past with its Facebook Messenger bot. With examples of Indian companies from other sectors, most notably Flipkart, highlighting the importance of inclusion of AI in their businesses, companies from the food industry may also push the country’s drive towards AI adoption. Google’s Association With AI Startups In India Was At An All-Time High In 2017 The shift towards an Artificial Intelligencecentric model of delivering services is soon becoming the norm of many organisations and Google is leading the way along with its contemporaries. The global search giant who is looking at an ‘AI first’ world, has been working towards that end through acquisitions and by extending its support to AI startups. Google’s Launchpad Accelerator programme proves that the company takes its push towards an AI first world very seriously. The programme boasts of a portfolio of startups across various countries in Asia, Africa, Europe and Central/South America.With many Indian AI-startups falling under Google’s radar, the companies push towards an ‘AI first’ world could see a significant contribution coming from the country. Souma Das Appointed AsThe Managing Director Of Teradata’s India Operations
www.quantinuumsimsr.com
Quriosity | Vol. 09, Issue 01 | January 2018 In a recent announcement by Teradata Corp., leading data and analytics company, it has appointed Souma Das as the Managing Director for its India operations. As per the statement released by the company, he will be responsible for providing leadership and overall strategic direction to the company’s India business overseeing field operations that include sales, customer management, marketing, professional services and customer support. Mobile Analytics Startup CleverTap Receives Fresh Funding From Recruit Holdings Co. Ltd As a part of its fundraising drive, analytics startup CleverTap has secured an undisclosed amount in funding from Japan-based Recruit Holdings Co.,Ltd. Recruit Holdings Co.,Ltd, provider of diverse consumer and business services, made the investment through its investment subsidiary, RSP India Fund LLC, an equity wing created in 2015 that funds Indian startups It helps firms understand customer behaviour and improves their customer retention, and caters to customers across different stages of development such as startups, SMBs and Fortune 500 companies. Its clientele includes global brands such as Samsung, Sony and McDonald’s.
Quant Fun
Q. Dr.Haathi has taken some loan from State bank of India at 8% interest rate per annum.He has also borrowed some money from ICICI at 10% interest.The total loan amount is Rs.10,000. He pays total Rs.950 as interest rate at the end of first year.How much money did he borrow from the State bank of India? Q. Janice left her home at 11 am, traveling along route 1 at 30 mph. At 1 pm, her brother Richard left home and started after her on the same road at 45 mph. At what time did Richard catch up to Janice???? Q. What is the total number of ways in which 55 chocolates can be divided among 5 boys such that each boy gets an odd number of chocolates? Q. A person has to climb 10 steps...He climbs in either a single step or 2 steps.In how many ways can he do it? Q. N=122333444455555.. and so on till 100 terms divided by 16 what is the remainder? Can you solve the following sudoku puzzle?
Indian Researchers Develop Light ScatterBased Algorithm To Detect Cancer In a path-breaking innovation that can revolutionize cancer treatment, a team of Indian researchers have developed an AIbased algorithm that can do more than just detect the disease. Researchers from the Indian Institute of Science Education and Research, Kolkata (IISER-K) and Indian Institute of Technology, Kanpur (IIT-K) have developed an algorithm based on the light scattering properties of healthy and precancerous cells.
7
www.quantinuumsimsr.com
Quriosity | Vol. 09, Issue 01 | January 2018
Quant Connect
Quantinuum, the Quant and Analysis committee of KJ Somaiya Institute of Management & Research aims to empower students and professionals alike to organise and understand numbers and in turn, to make good and rational decisions as future managers. the newsletter published monthly, consists of a gamut of articles for readers ranging from beginners to advanced learners so as to further enrich the young minds understand the contributions made to the field of mathematics along with a couple of brain-racking sessions of puzzles to tickle the gray cells. For any further queries and feedback, please contact the following address: KJ Somaiya Institute of Management Studies and Research, Vidya Vihar, Ghatkopar East, Mumbai- 400077 or drop us a mail at newsletter.quantinuum @gmail.com
Mentor: Email:
Prof. N. S. Nilakantan (+919820680741) nilakantan@somaiya.edu
Team Leaders: Vaibhav M (+917708521382) Maheshwaran Kumar (+919566173411) Rishita Shah (+919867290018)
8
www.quantinuumsimsr.com