5 stars based on
What is the price of an orange? As a starting point, one would derive the price of an orange based on two things: If John wants to sell it for USD2. Or if there is a drought the supply of oranges will become less, so more people are trying to buy less oranges, which can also drive the price up.
The origins of a trustless system. Bitcoin as a payment system Point and figure chart bitcoin value for consumers and merchants. Bitcoin as digital gold A new digital asset class to consider. Bitcoin as the internet Global, open to all for innovation and use.
How do I get Bitcoin? Learn about earning, trading, buying and mining. How is the price of Bitcoin calculated? A primer on supply and demand. Is Bitcoin used by criminals?
What is Bitcoin mining? What are Bitcoin private keys? How do I protect my private keys? Wallets, vaults and private keys. How do I keep my Bitcoin safe? Recognising scams to reduce risks. What are the risks with Bitcoin? With potential comes risk. The first global computer. Where did Ethereum come from? The arrival of a more versatile blockchain. Distinguish between Bitcoin and Ethereum. What are smart contracts? A way to connect, automate and eliminate inefficiency.
How do I get Ethereum? You can earn, buy or mine Ether. How is the price of Point and figure chart bitcoin value calculated? Supply and demand, with some context. Is Ethereum used by criminals? What is Ethereum mining?
Transaction processing, for a fee. What are Ethereum private keys? These give you the right to spend your digital currency. How do I keep my Ether safe? What are the risks with Ethereum? Time series analysis is not an easy topic I must say. At least for me. The purpose of this post is to reinforce what I have learned by implementing and trying to explain to others. Time series data should be treated differently to other types of data.
One of the statistical data assumptions is its independence. Independence means the value of one observation does not influence or affect the value of other observations. But in time series data, each data point is close together in time, and they are not completely independent of their adjacent values. So we need a different approach to model time series data. First, start by loading dependencies. For this post, I will extract data from Yahoo Finance. As you can see, the data has six columns.
As the name suggests, rather than calculating the average on the whole dataset, moving average also called rolling mean calculates the average of a subset with a certain window size, and shifts forward. Moving average is used to smooth out short-term fluctuations and highlight longer-term trends or cycles. I chose the 4th quarter of to plot where there is a strong trend in the data to clearly see how moving average works. Compared to the original observation, which is plotted with a blue line, we can see the curve of the lines get smoother as window sizes get bigger.
Another use case of moving average is in a trading strategy called dual moving average crossover. I guess by changing the time frames shorter, the strategy will be more short-sighted. But I am not a finance expert, and I am not sure if I am making any fundamental mistakes by changing the time frame shorter.
The concept of a dual moving average crossover is fairly straightforward. Calculate two moving averages of the price, one average would be the short term and the other long term. The long term moving average will have a lower variance and will move in the same direction as the short term moving average but at a different rate. The different rates of direction induces points where the values of the two moving averages may equal and or cross one another.
These points are called the crossover points. In the dual moving average crossover trading strategy, these crossovers are points of decision to buy or sell the currencies. However, while looking for more material on the subject, I have also found out that there is an opposite approach to the same signal points. The above method introduced by the author is called The Technical Approach and the other approach is called The Value Approach.
The Value Approach claims that when the short-term average crosses from below to above the long-term average, that the investment is now overvalued, and should be sold. It is very interesting to see that there are two contradictory views on the same situation. But in this post, I will focus on The Technical Approach. With The Techincal Approach, I would have started my investment with 1, It might be a naive calculation, not taking into account transaction fees or other costs which might have occurred.
But in a simplified calculation, not a bad investment. If The Value Approach is applied to the same graph, the situation might not be the same.
Selling short is different than selling an asset we already own, which is called selling long. In simple words, recent prices are given more weight than past prices, and the degree of the contribution exponentially decay as the time period goes further to the past from the current observation. Here alpha is the decaying factor. Maybe there is a standard that everyone agrees even without explicitly stating it, but I have spent a good amount of time trying to figure this out, but I am still not sure.
If you are familiar with this, any help would be appreciated. I have found the answer to my own question above. There are four different ways you can specify it. You can also directly specify alpha from Pandas version 0. But intuitively, compared to simple moving average, exponential moving average will react faster to the more recent movement of the price by giving more weights to the current values.
Only this time we look for the crossing points between the original observation and EMA. With EMA trading strategy, I would have started my investment with 4, It was a fun toy project to actually implement what I have learned during the class, and I will try to explore further with time series analysis and Bitcoin data.
But should I invest in Bitcoin? Thank you for reading and you can find the Jupyter Notebook of the above code from the below link. It is an attempt to combine theory, applications, and numerical methods, and cover each of point and figure chart bitcoin value fields with the same weight. As a result, a lot of hasty market participants pick up the initial spurt, start trading and already stimulate traffic on their own.