Sadness Quantification

It is almost midnight here, the time of the day when I lose every common sense in order to indulge myself in thoughts. These days, all of my thinking is about sadness. Surrounded by such a melancholy ambiance, my heart is suffocated in these invisible walls of strangers and acquaintances. If only there was a cure for this mental sickness, ... if and only if the sadness can be estimated, measured and quantified via a math formula. 

Is this a kind of depression that could last for years? Or is it just another temporary mood swing? The easiest variable would be “t”, the duration of sadness feeling. The longer I feel sad, the bigger that sadness is. “t” could be seconds, minutes, hours, dates, or even years. There are measurement errors such as we did know when the sadness started, or which indicates the end of sadness feeling. A depressed person could both laugh so hard and still carry a heavy heart of sorrow. The complicated part is that no one knows the value of “t” until the sadness is over. However, feelings could be infinite, but our human lifetime is so finite, which make it harder to know whether “t” is bounded or converges to a specific value. We can only make assumptions on max(t)=lifetime and min(t)=0. 

On the other hand, a forgetful one could smile away troubles, but quickly be reminded of the sadness after that. Therefore, “t” could either be “continuous” or “discrete”. Is “t” an independent variable? Or is it correlated with another one? For instance, the “origin” of sadness, or the “personality” of the sadness bearer could directly affect “t”. Also, let’s consider “t” as a stopping time, as Buddha teaches us “People will eventually let go once it hurts too much”. Then we need the stop value, which is the immeasurable “tolerance”. Up to this point, “t” is measurable only if we use approximation “lim” under “subjective probability”. 

The most important causal variable is no other but the “origin” of sadness. Only to list a few, we already have tons of reasons to feel sad in this life: break-up, fail the exam, lose the job, grieving, etc. It would be much simpler if we could put all the causes in a prioritized order, and give a specific value to each one, and then we will know the most heart-breaking ones to avoid. Unfortunately, this priority list is not deterministic. It varies on different periods of life, and on different kinds of “personality”. What if we only consider the “origin” with binary value? Someone once told me the sadness from something that we regret doing or not doing is much greater than the one that other people do on ourselves. Will 0 and 1 be justified for every case? If a couple were separated, who would feel sadder? Is it the one who said the break-up word, or the one who accept that? 

Speaking about the people, there will always be someone in the “origin” of sadness and someone else who act as sadness bearer. As the subject here is the sadness bearer, the “personality” of he or she will be taken into account. This could be modeled with binary value as strong people often overcome sadness better and quicker than people with soft “personality”. However, even the strongest man on Earth has weaknesses and soft points. If the sadness is from someone we love, this could have a much greater impact than the sadness from an acquaintance. An accident happened to us or to some specific people we know would cause more sadness? Another great thing that should be included in this model is the sadness “killer”. What would cheer us up on a blue day? If the sadness can be killed easily, than the significance of it would be much lower than expected. Sadness “killer” parameter could be the only one that has negative value in this formula (and ironically has positive impact in real meaning). Personally, my “killer” would be 3F: family, friends, and food. However, how can I know which “killer” could be strong enough to transform this sadness formula into a zero equation? Subjective again? Or should I just try to stay focused, be positive, collect happiness and then one day, all of my sadness will be equalized? 

Somehow, now I want to research on happiness quantification more than sadness. Or should I have a complex formula that works for both sadness and happiness? A Mixture Model for feelings? Maybe another time ... since next time, it will be all about EM Algorithm!